PCSK9 Inhibitors Do Not Increase Cognitive Risk

PCSK9 Inhibitors Do Not Increase Cognitive Risk | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article PCSK9 Inhibitors Do Not Increase Cognitive Risk Dongsheng Ni, Zeying Feng, Dehuan Liang, Zhaolai Qi, Yao Lu, Liyong Zhu, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8520824/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Proprotein convertase subtilisin–kexin type 9 (PCSK9) inhibitors are potent lipid-lowering therapies, yet concerns regarding their neurocognitive safety persist amidst conflicting evidence. Methods We applied an integrated approach combining real-world pharmacovigilance, genetic epidemiology, and preclinical models. Disproportionality analysis was performed using FDA Adverse Event Reporting System (FAERS) data to assess cognitive adverse events. Two-sample Mendelian randomization (MR) was used to analyze the causal effects of apolipoprotein B (ApoB)-lowering and PCSK9 inhibition on Alzheimer’s disease (AD) via UK Biobank/FinnGen data, with a factorial MR design to evaluate interactions stratified by ApoB/PCSK9 polygenic risk. Preclinically, hepatocyte-specific knockout and pharmacological inhibition of PCSK9 in AD-transgenic mice were assessed through cognitive testing and neuropathological evaluations. Results No increased risk signal for cognitive events was identified with PCSK9 inhibitors in FAERS analysis (AD Reporting Odds Ratio [ROR] = 0.62, 95% CI 0.45–0.87; dementia ROR = 0.92, 0.79–1.08), unlike statins which showed a significant signal (AD ROR = 2.86). MR analysis revealed no causal association between PCSK9 inhibition (OR = 1.00, 0.93–1.07) or its ApoB-lowering effects (OR = 0.79, 0.56–1.11) and AD. In AD-transgenic mice, PCSK9 knockout reduced low-density lipoprotein-cholesterol by 54% without impairing cognition or neuron integrity, or increasing amyloid-β plaques and astrocytes. Pharmacological inhibition of PCSK9 similarly showed no cognitive deficits. Conclusions Integrated evidence from pharmacovigilance, genetic epidemiology, and preclinical models robustly supports the neurocognitive safety of PCSK9 inhibitors, including in individuals with high ApoB levels. These findings affirm their long-term neurocognitive safety profile in the management of atherosclerotic cardiovascular disease. Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 INTRODUCTION Atherosclerotic cardiovascular disease (ASCVD) remains the leading cause of global morbidity and mortality[1,2], underscoring the imperative for aggressive dyslipidemia management[3–5]. This has driven the development of progressively potent low-density lipoprotein (LDL) cholesterol-lowering therapies. However, given the indispensable role of cholesterol in myelin synthesis and neuronal membrane integrity, the pursuit of profound lipid reduction has simultaneously raised concerns regarding potential neurocognitive adverse events[6,7]. These concerns initially surfaced with statin therapies, often on the basis of observational data and spontaneous reports[8–10]. Given the capacity of proprotein convertase subtilisin–kexin type 9 (PCSK9) inhibitors to achieve even more profound reductions in atherogenic lipoproteins than statins do, the emergence of these agents has logically intensified such concerns. Indeed, while initial studies have begun to address this issue, the findings remain controversial. For example, the Open-Label Study of Long-Term Evaluation Against LDL-C (OSLER-1 and OSLER-2) trials noted a numerically higher incidence of adverse events with evolocumab than with standard therapy[11]. Conversely, dedicated neurocognitive trials such as EBBINGHAUS, which prospectively assessed cognitive changes as a primary endpoint, reported no adverse effects with evolocumab[12]. A critical limitation of existing neurocognitive safety studies is the general omission of baseline apolipoprotein B (ApoB) levels as a potential modulating factor. ApoB, the primary structural apolipoprotein of all atherogenic lipoproteins (LDL, very low-density lipoprotein, intermediate-density lipoprotein, and lipoprotein a), provides a direct measure of the total atherogenic particle burden[13,14]. This oversight is particularly salient given the emerging links between ApoB metabolism and cognitive function[15]. It is plausible that individuals with a high baseline ApoB burden—representing a potentially ‘vulnerable brain’ phenotype—might be more susceptible to any subtle neurocognitive perturbations[16,17]. Furthermore, these patients often require more intensive or prolonged lipid lowering therapy (including PCSK9 inhibitors) to reach guideline-recommended targets, resulting in greater cumulative drug exposure[18]. Consequently, a compelling need exists to delineate the long-term cognitive impact of PCSK9 inhibitors, specifically in patients with high baseline ApoB concentrations. However, prospectively investigating this topic in a dedicated large-scale, long-term randomized controlled trial (RCT) presents substantial logistical and ethical hurdles. Therefore, a strategy integrating evidence from diverse, complementary methodologies is warranted to assess potential risks in this vulnerable subgroup. In the present study, we aimed to synthesize and evaluate evidence from: 1) analysis of spontaneous adverse event reports from a global pharmacovigilance database; 2) Mendelian randomization (MR) studies examining genetically proxied PCSK9 inhibition stratified by genetic predisposition to elevated ApoB levels; and 3) findings from preclinical animal models mimicking high ApoB states with concomitant PCSK9 inhibition. By triangulating insights from these approaches, we sought to provide a more comprehensive assessment of the cognitive safety of PCSK9 inhibitors, particularly in the context of a preexisting high ApoB particle burden. RESULTS Disproportionality analysis using the FAERS We conducted a pharmacovigilance study of adverse drug reactions (ADRs) associated with PCSK9 inhibitors and statins using data from the FDA Adverse Event Reporting System (FAERS) database, a publicly accessible repository of safety reports submitted by patients, healthcare professionals, and pharmaceutical companies. A total of 55,357,463 individual case safety reports (ICSRs) were collected from the FAERS database between the first quarter of 2004 and the fourth quarter of 2024. The specific generic drug names for Statins are detailed in Supplementary Table 1. PCSK9 inhibitors include evolocumab and alirocumab. Adverse event list is provided in Supplementary Table 2. Among these, 525,251 reports listed statins as a suspected drug. Of these statin-associated reports, 229 were linked to the MedDRA High-Level Term (HLT) Alzheimer’s disease (AD) (incl subtypes)’ and 383 to ‘Dementia (excl Alzheimer’s type)’. Similarly, 371,200 reports involved PCSK9 inhibitors as a suspected drug, with 36 reports corresponding to AD (incl subtypes)’ and 167 to ‘Dementia (excl Alzheimer’s type)’. In our disproportionality analysis, no significant signal for an increased risk of cognitive adverse events was detected for PCSK9 inhibitors. For ‘Alzheimer’s disease (incl subtypes)’, the ROR was 0.62 (95% Credibility Interval (CI), 0.45–0.87) and the corresponding Information Component (IC) was − 0.68 (95% Credibility Interval, -1.14 to -0.19). For ‘Dementia (excl Alzheimer’s type)’, the ROR was 0.92 (95% CI, 0.79–1.08) with an IC of -0.11 (95% Credibility Interval, -0.33 to 0.11). Conversely, statin use was associated with a significant disproportionality signal for cognitive decline across both analytical methods. Specifically, the Reporting Odds Ratio (ROR) for ‘AD (incl subtypes)’ with statin use was 2.86 (95% CI, 2.51–3.26) with an IC of 1.49 (95% CI, 1.29 to 1.67), and for ‘Dementia (excl Alzheimer’s type)’ the ROR was 1.51 (95% CI, 1.36–1.67) with an IC of 0.58 (95% CI 0.43 to 0.73) (Fig. 1 ). MR analysis Two-sample MR Supplementary Table 3 lists the genetic markers employed for constructing ApoB and genetic PCSK9 scores. Supplementary table 4 presents comprehensive details concerning the final single nucleotide polymorphisms (SNPs) for ApoB and PCSK9, including effect and other alleles, beta coefficients, standard errors, and corresponding P-values. The F-statistics for the chosen SNPs varied between 21.52 and 28.67 for ApoB, and from 32.07 to 2288.71 for PCSK9, which suggests an absence of weak instrumental variable bias. In the univariable two-sample MR analysis assessing the causal effect of ApoB on AD, the IVW method showed no statistically significant association [odds ratio (OR) = 0.79, 95% CI = 0.56 to 1.11, P = 0.17] (Fig. 2). Consistent with this finding, supplementary MR approaches, including the weighted median, MR-Egger, and maximum likelihood methods, yielded similarly non-significant estimates (all P > 0.05), suggesting a lack of robust evidence for a causal role of ApoB in AD risk. Meanwhile, the identical approaches utilized to explore the causal effect of PCSK9 on AD also did not yield statistically significant results as shown in Fig. 2 (all P > 0.05). Figure 2 . Forest plot of MR analysis of ApoB and PCSK9 on AD risk. Abbreviations: ApoB, apolipoprotein B; CI, confidence interval; MR, Mendelian randomization; OR, odds ratio; PCSK9, proprotein convertase subtilisin/kexin type 9. Sensitivity analyses The Cochran's Q test and MR-PRESSO analyses indicated no significant heterogeneity or outlying variants in ApoB. The Cochran's Q test indicated significant heterogeneity in PCSK9, yet MR-PRESSO failed to identify any abnormal SNPs. This observation suggests that the observed heterogeneity may be attributable to the weak pleiotropy of the wide distribution or minor bias in the multi-instrumental variables (Supplementary Table 5). The primary analyses were conducted employing a random-effects inverse variance-weighted (IVW) model, and the findings were consistent with robust methods (e.g., weighted median), thereby substantiating the reliability of the conclusions. The MR‒Egger regression intercept terms both exceeded the significance threshold (P > 0.05), confirming minimal horizontal pleiotropic effects. Initial visual inspection of forest and funnel plots (Supplementary Figs. 1–4) supported these findings. Additional leave-one-out sensitivity tests (Supplementary Figs. 5–6) further confirmed the stability of our primary outcomes. For comprehensive evaluation, we implemented four distinct analytical approaches and generated scatter plots depicting AD associations (Supplementary Figs. 7–8). 2×2 factorial MR analysis Of the initial 501,938 participants from the UK Biobank, 2,653 were excluded due to a diagnosis of AD or stroke at or before baseline. An additional 141,23 individuals were excluded for missing genetic score data, and 97,782 were excluded due to incomplete covariate information. The final analytical sample comprised 387,380 participants, among whom 184,612 (47.66%) were male, with a median age of 57 years (Interquartile Range (IQR): 50–63). Over a median follow-up of 13.64 years (IQR: 12.91–14.35), a total of 2,914 incident AD cases were documented. In the 2×2 factorial MR analysis, using the “High genetic ApoB score & Low genetic PCSK9 score” group as the reference category to directly test our primary hypothesis, none of the comparison groups demonstrated statistically significant differences in AD risk (Table 1 ). These findings align with our primary MR analyses, which similarly failed to support a causal role for either biomarker individually. Table 1 Combined Genetic Predisposition to ApoB and PCSK9 and Alzheimer’s Disease Risk Model High ApoB_PRS & Low PCSK9_PRS (HR, P) Low ApoB_PRS & Low PCSK9_PRS (HR, P) Low ApoB_PRS & High PCSK9_PRS (HR, P) High ApoB_PRS & High PCSK9_PRS (HR, P) Crude Model ref, - 0.96 (0.87, 1.06), 0.45 0.95 (0.86, 1.06), 0.36 0.97 (0.87, 1.07), 0.50 Model 1 ref, - 0.97 (0.88, 1.08), 0.60 0.98 (0.88, 1.08), 0.65 0.98 (0.88, 1.08), 0.66 Model 2 ref, - 0.97 (0.88, 1.08), 0.60 0.97 (0.88, 1.08), 0.62 0.98 (0.88, 1.08), 0.65 Model 3 ref, - 0.97 (0.88, 1.07), 0.55 0.97 (0.88, 1.08), 0.57 0.97 (0.88, 1.08), 0.61 Model 1: Adjusted for age, sex, ethnicity Model 2: Adjusted for age, sex, ethnicity, Townsend deprivation index Model 3: Adjusted for age, sex, ethnicity, Townsend deprivation index, smoking status, drinking status, exercise, BMI, hypertension, diabetes, dyslipidemia Abbreviations: AD = Alzheimer’s disease; ApoB = Apolipoprotein B; BMI = Body Mass Index; CI = Confidence Interval; HR = Hazard Ratio; PCSK9 = Proprotein Convertase Subtilisin/Kexin type 9; PRS = Polygenic Risk Score; Ref = Reference. Animal experiment 1: hepatocyte-specific knockout of PCSK9 Cognitive impairment mouse line To investigate whether the inhibition of mouse PCSK9 induces neurocognitive dysfunction, we designed two mouse models for validation. APP/PS1, double transgenic mice expressing a chimeric mouse/human amyloid precursor protein (Mo/HuAPP695swe) and a mutant human presenilin 1 (PS1-dE9), is the most common neurodegenerative model in the early-onset AD [19,20]. Importantly, this mouse line begin to develop beta-Amyloid (Aβ) plaques and spatial memory deficit at 6-7 months of age[21]. Genetically, we bred hepatocyte-specific knockout (HKO) of PCSK9 based on the APP/PS1 mice. It’s suggested that PCSK9 in the brain may affect the cleavage of Aβ protein[22], and there are also reports that PCSK9 exacerbates the release of inflammatory factors from astrocytes[23]. Therefore, whole-body knockout of PCSK9 may introduce many interferences in determining whether peripheral PCSK9 intervention affects cognitive function, while PCSK9 HKO can overcome this deficiency. Learning and memory We raised APP/PS1-PCSK9 HKO mice under conventional breeding conditions until the age of 9 months to simulate long-term inhibition of PCSK9 (Figure 3A). Subsequently, we assessed whether long-term PCSK9 inhibition affected cognitive function using behavioral and neuropathological experiments (Figure 3A). The results of the western blotting (WB) experiment showed that compared with the control mice (APP/PS1-PCSK9 F/F ), the expression of PCSK9 in the livers of APP/PS1-PCSK9 HKO mice was significantly reduced, without affecting the expression in other tissues such as brain, kidney, heart, fat, and muscle (Supplementary Figure 9A). We also assessed the serum lipid profile alteration of hepatocyte-restricted PCSK9 deficiency in above mouse lines. Our results demonstrated that compared with control mice, APP/PS1-PCSK9 HKO mice exhibited approximately 54% reductions in average serum LDL-C (Supplementary Figure 9B), 31% reductions in average total cholesterol (TC) (Supplementary Figure 9C) and 32% reductions in average non- high-density lipoprotein cholesterol (non-HDL-C) (Supplementary Figure 9D). The water maze results indicated that as the number of training days increased, escape latency taken by both APP/PS1-PCSK9 F/F and APP/PS1-PCSK9 HKO mice to locate the submerged platform gradually decreased, with no statistically significant differences between groups (Supplementary Figure 9E). During the probe trial of water maze (Figure 3B), the two groups of mice were no statistically significant differences in the escape latency (Figure 3C), the number of platform crossings (Supplementary Figure 9G), and the time spent in the target quadrant (Supplementary Figure 9H), with the similar average swimming speed (Supplementary Figure 9F). The Y-maze alternation test results showed, under the condition of the same total number of entries into the three arms of the Y-maze (Supplementary Figure 9I), there were no statistically significant differences in the proportion of consecutive entries into the three arms (Figure 3D, E) between the two groups of mice. The pathological staining results of the brain tissue of the mice indicated that hepatocyte-specific knockout of PCSK9, did not affect the size or number of Nissl bodies (Figure 3F), Aβ + plaques (Figure 3G), or GFAP + astrocytes (Figure 3H) in the hippocampal region. Nerve and protection behavior In addition to direct assessment of learning and memory, we also examined cognition-related phenotypes, including locomotor activity, nerve-like behavior, and motor coordination. In the open field test (Figure 4A), APP/PS1-PCSK9 HKO and APP/PS1-PCSK9 F/F mice exhibited comparable numbers of entries (Figure 4C) into and times spent in the central zone (Figure 4D) while demonstrating similar total travel distances (Figure 4B). During the rotarod test, the proportion of mice falling from the rotating rod increased with escalating rotation speed in both APP/PS1-PCSK9 HKO and APP/PS1-PCSK9 F/F groups, with no significant intergroup differences observed in fall percentage (Figure 4E). Furthermore, the number of falls within the 300-second test period was also comparable between groups (Figure 4F). Animal experiment 2: neutralization of PCSK9 pharmacologically Human- like lipid profile mouse line Mice are naturally not susceptible to ASCVD, and one potential reason is that the serum lipid profile of mice differs significantly from that of humans, which is characterized mainly by a lower ratio of LDL to HDL fraction. This may lead to potential inconsistencies in the results of PCSK9 inhibition on cognitive function between mice and humans. However, it has been reported that ApoB transgenic mice can significantly increase the ratio of the LDL to HDL fraction, resulting in a human-like serum lipid composition[24,25]. Moreover, the ApoB gene is a common causative gene leading to hypercholesterolemia[26,27]. Third, increasing evidence has shown that ApoB is the most accurate marker of ASCVD risk [13]. Fourth, PCSK9 mAbs usually rapidly decrease elevated LDL-C, which is different from the findings in the PCSK9 knockout model. Considering these points, we designed a second mouse model to explore the impact of PCSK9 inhibition on cognitive function. We bred APP/PS1-ApoB triple transgenic mice, fed them a high-fat high-cholesterol diet (HFHCD) to induce hypercholesterolemia. And then administered a PCSK9 mAb to observe whether a rapid reduction in LDL-C would impair the cognitive function of APP/PS1-ApoB mice, which exhibited a human-like lipid profile (Supplementary Figure 10A). Cognitive function Our data revealed that the HFHCD increased mouse body weight (Supplementary Figure 10B). Then PCSK9 mAb sharply decreased the serum ApoB content (~66%; Supplementary Figure 10C), LDL-C content (~43%; Supplementary Figure 10D), TC content (~32%; Supplementary Figure 10E) and nonHDL-C content (~37%; Supplementary Figure 10F). Further water maze (Supplementary Figure 10G-K), Y-maze (Supplementary Figure 10L-N), open field test (Supplementary Figure 10O-R), rotarod test (Supplementary Figure 10S-T) and brain tissue pathological staining (Supplementary Figure 10U-W) results revealed that pharmacological neutralization of PCSK9 did not alter locomotor activity, nerve-like behavior, motor coordination or balance in APP/PS1-based mice. Therefore, neither the hepatocyte-specific genetic knockout of PCSK9 nor its pharmacological neutralization adversely affected cognitive function in APP/PS1 mice. DISCUSSION and CONCLUSION PCSK9 is a powerful drug target for therapeutic strategies aimed at lowering LDL-C, which regulates LDLR turnover of hepatocytes[28,29]. However, the impact of PCSK9 on cognitive function remains controversial, as evidence from both clinical and preclinical studies is inconsistent [11,12,23,30]. PCSK9 is also known to influence central nervous system development, neuronal differentiation, and apoptosis[12,22,23]. Furthermore, the Open-Label Study of Long-Term Evaluation against LDL Cholesterol (OSLER) study revealed that neurocognitive events were reported more frequently in the evolocumab group an assessment of 11.1 months (median)[11]. These uncertainties highlight the need for rigorous investigation into the potential neurocognitive effects of prolonged PCSK9 inhibitor use. To address the issue of PCSK9 inhibitors on cognitive safety, we conducted the first study to systematically combine pharmacovigilance, genetic epidemiology, and preclinical evidence in a single investigation. The present study did not find evidence of adverse cognitive-related side effects related to the inhibition of PCSK9, in concurrence with earlier safety assessments based on short-term clinical studies and genetic studies[12,30–32]. In our disproportionality analysis of FAERS data, no significant association was detected between the use of PCSK9 inhibitors and increased reporting of adverse events related to cognitive function. However, the use of statin has (Figure 1). To our knowledge, we provides the longest pharmacovigilance follow-up to date (~20 years)[33]. MR is a methodological approach in human genetics that facilitates causal inference. It effectively reduces the confounding bias frequently observed in conventional epidemiological studies. When the assumptions underlying MR are rigorously met, the inherent randomization of genetic instruments offers a framework akin to a RCT, allowing for the assessment of long-term causal effects of exposures[30]. In the univariable two-sample MR analysis, we cannot conclude a causal relationship between PCSK9 and AD from the UKB source by 4 MR analysis approaches (Figure 2). Furthermore, our data also failed to support a causal role of combined genetic predisposition to ApoB and PCSK9 and AD risk by using 2×2 factorial MR analysis (Table 1). Except for real world data and genetically proxied PCSK9 inhibition, we also developed two gene-edited animal models to verify whether the inhibition of PCSK9 impairs cognitive function. One is 9 mo APP/PS1-PCSK9 HKO mice, which are used to simulate long-term inhibition of PCSK9 (Figure 3). The other group included PCSK9 mAb-administered APP/PS1-ApoB triple transgenic mice to simulate short-term inhibition of PCSK9 (Supplementary Figure 10). Our mouse behavior tests and brain pathological staining indicated that repression of PCSK9 did not exacerbate cognitive impairment, genetically or pharmacologically. A recent study revealed that PCSK9 ablation attenuates Aβ pathology, neuroinflammation and cognitive dysfunctions in 5XFAD mice[23]. However, they used PCSK9 whole-body knockout mice, in which brain PCSK9 was also deleted. This differs from approved therapeutics that use PCSK9 mAbs (which were also employed in our animal experiments), as mAbs typically cannot cross the blood‒brain barrier to directly affect cerebral PCSK9 function. The currently approved PCSK9 siRNA drugs specifically target the liver and do not directly affect brain PCSK9 expression[34,35]. ApoB combines LDL-C as a validated therapeutic target in dyslipidemia[36], although its neuroprotective threshold remains debated[16,25]. Our preclinical data confirmed that aggressive lowering of ApoB (mean reduction: 66%) in AD-relevant mouse models induced no detectable neurocognitive deficits (Supplementary Figure 10). For individuals with ApoB mutation-induced hypercholesterolemia, PCSK9 inhibitors do not pose additional cognitive risks when used to lower cholesterol (Figure 2 and Supplementary Figure 10). As far as we known, no previous study has examined this specific population using genetic and preclinical approaches. Our study is subject to several limitations. First, the study populations were exclusively of European or American ancestry. Consequently, our findings may not be generalizable across all populations. Also, an APP/PS1 transgenic mouse model may not fully replicate human cognitive disease pathophysiology. Second, there were instances of heterogeneity within the MR analysis. The reliance on GWAS data precludes the exploration of potential nonlinear relationships or stratification effects that may vary according to age, health status, or sex. Finally, the animal experiments did not include female mice, which may limit the applicability of the results. By integrating real-world adverse drug reaction data, MR studies, and animal experimental results from this study, combined with existing related research, we conclude that, compared with statins, PCSK9 inhibitors do not increase the risk of neurocognitive impairment, especially given their cardiovascular benefits. This conclusion also applies to patients with hypercholesterolemia caused by ApoB mutations (Figure 5). Besides, these integrated approaches establish a new standard for drug safety evaluation that could be applied to other therapeutic areas. MATERIALS AND METHODS Study Design To address the issue of PCSK9 inhibitors on cognitive safety, pharmacovigilance analysis, MR analysis and preclinical gene-editing mice were conducted. Pharmacovigilance data analysis FAERS data files spanning from the first quarter of 2004 to the fourth quarter of 2024 were downloaded for this analysis. Adverse event terms within the FAERS database were coded via the most current version of the Medical Dictionary for Regulatory Activities (MedDRA), version 27.1. Drug Definitions The WHO Drug Dictionary was utilized to standardize all drug names within the database. Statins were defined via the Anatomical Therapeutic Chemical (ATC) classification code C10AA; the specific generic drug names included are detailed in Supplementary Table 1. PCSK9 inhibitors include evolocumab and alirocumab. Adverse Event Definition Cognitive-related adverse events were defined via standardized MedDRA terminology at the HLT level, encompassing the HLTs AD (incl subtypes)’ and ‘Dementia (excl Alzheimer’s type)’. A comprehensive list of the specific Preferred Terms (PTs) included under these HLTs is provided in Supplementary Table 2. Statistical analysis in FAERS Disproportionality analysis is a fundamental data mining and signal detection technique —a statistical method used to identify potential safety signals—in pharmacovigilance and pharmacoepidemiology aimed at identifying statistically significant safety signal associations between drugs and adverse events (AEs) within spontaneous reporting system databases. In this study, we employed the reporting odds ratio (ROR) and the Bayesian Confidence Propagation Neural Network (BCPNN) for disproportionality analysis to ensure methodological robustness. A signal was considered statistically significant if the lower limit of the 95% CI for the ROR was greater than 1. For BCPNN, a signal was identified if the lower bound of the 95% credibility interval for the IC was greater than 0. All analyses required that the number of ICSRs for the drug‒event combination was three or more[28]. Mendelian randomization analysis Genome-wide association studies (GWAS) data sources The GWAS data of ApoB and PCSK9 were sourced from UKB (n = 46,218, aged 40 to 69 years old)[29]. The summary genetic data for AD were acquired from FinnGen (version R10)[30] including 10,520 cases and 401,661 controls of Finland descent (which are accessible at https://www.finngen.fi/en). Consent Ethical oversight for the UKB was provided by the Northwest Multi-Center Research Ethics Committee, with all study participants providing written consent. Our research team gained authorization to use these resources after receiving approval from the UKB Ethics and Governance Council (Application No. 75283). SNP selection To investigate the potential causal effects of ApoB and PCSK9 on AD susceptibility, rigorous instrumental variable selection criteria were implemented through multiple quality control steps. The initial stage involved identifying genetic variants exhibiting associations with the exposure variables, where single nucleotide polymorphisms (SNPs) demonstrating statistical significance with ApoB and PCSK9 were selected as candidate instruments. Due to the absence of genome-wide significant SNPs (P < 5 × 10 −8 ) for ApoB, a more lenient threshold (P < 5 × 10 −6 ) was adopted to ensure sufficient instrumental variables (IVs) for robust analysis. Subsequently, we clumped SNPs based on the basis of the European 1000 Genomes Project reference panel (r 2 10000 kb) to guarantee genetic independence among selected variants. We applied a minor allele frequency threshold of > 0.01 to exclude rare variants and ensure genotype reliability. Data harmonization was then conducted to align effect alleles across exposure-outcome datasets. Finally, instrument strength was evaluated through F-statistics, with values less than 10 suggesting potential weak instrument bias and necessitating exclusion from further MR analysis. Two- sample MR We employed multiple MR approaches, including inverse variance-weighted (IVW) regression, weighted median estimation, MR‒Egger regression, and maximum likelihood methods. The IVW method served as our primary analytical framework, operating under the assumption that all selected genetic variants meet instrumental variable assumptions[31]. The weighted median estimator provides robust causal inference when at least 50% of the weighting derives from valid instrumental variables[32]. MR‒Egger regression offers additional protection against directional pleiotropy by allowing all genetic variants to exhibit some pleiotropic effects while providing valid causal estimates[33]. For scenarios involving weak instrumental variables, the maximum likelihood method was implemented to ensure accurate confidence interval estimation[34]. Comprehensive methodological details for these analytical techniques are available in previously published literature [35,36]. Sensitivity analyses To ensure the robustness of our findings, we implemented multiple sensitivity analyses. First, we assessed heterogeneity across genetic variants via Cochran's Q test [37]. The MR-PRESSO method was subsequently applied to detect and correct for horizontal pleiotropy by identifying and removing outlier variants that might bias the causal estimates[32]. When the global test indicated significant pleiotropic effects (P < 0.05), we excluded the identified outliers and repeated the IVW analysis with the remaining SNPs. Second, we evaluated potential directional pleiotropy through MR‒Egger regression, interpreting an intercept term with P > 0.05 as evidence against substantial horizontal pleiotropy[38]. Finally, we conducted leave-one-out analyses to examine whether individual SNPs disproportionately influenced the overall results, thereby identifying potential pleiotropic effects attributable to specific genetic variants. 2×2 factorial MR analysis The present study utilized data obtained from the UKB, which requires the formal approval of research proposals prior to data access. Ethical oversight for the UKB was provided by the Northwest Multi-Center Research Ethics Committee, with all study participants providing written consent. Our research team gained authorization to use these resources after receiving approval from the UKB Ethics and Governance Council (Application No. 75283). The genetic markers employed for constructing ApoB and genetic PCSK9 scores are comprehensively listed in Supplementary Table 3, while the variant selection methodology has been previously detailed in published works[39,40]. Elevated scores correspond to increased genetic predisposition for elevated circulating protein concentrations. We performed 2×2 factorial MR analysis, excluding participants with an AD or stroke diagnosis at or before baseline and participants missing covariates. Then, we split them to 4 groups based on the median of genetic ApoB and PCSK9 score. Critically, to address our primary research hypothesis regarding the potential vulnerability of individuals with high baseline ApoB burden to neurocognitive effects from intensive PCSK9 inhibition, we specifically designated the “High genetic ApoB score & Low genetic PCSK9 score” group as our reference category. Our models were adjusted for age, sex, ethnicity, Townsend deprivation index, smoking status, drinking status, exercise, BMI, hypertension, diabetes, dyslipidemia. Animal experiments Mouse lines APP/PS1 mouse (MMRRC Strain #034832-JAX), double transgenic mice expressing a chimeric mouse/human amyloid precursor protein (Mo/HuAPP695swe) and a mutant human presenilin 1 (PS1-dE9), were obtained from the Shanghai Model Organisms Center, Inc. The PCSK9 flox/flox mouse strain (S-CKO-00068) and Alb-Cre (C001006) mouse strain, which expresses Cre recombinase under the albumin promoter, were purchased from Cyagen Biosciences (Suzhou, China). Tg(ApoB)1102Sgy N20+ mice (Model No. 1004, commonly referred to as ApoB mice), which exhibit high levels of human ApoB100, were purchased from Taconic Biosciences, Inc. APP/PS1-PCSK9 HKO mice were generated by multiple crossings among APP/PS1 transgenic mice, PCSK9 Flox/Flox mice and Alb-Cre mice. The resulting offspring with the genotype APP/PS1-PCSK9 F/F -Alb-Cre + (named APP/PS1-PCSK9 HKO ) exhibited hepatocyte-specific knockout (HKO) of PCSK9 and overexpression of human APP/PS1. APP/PS1-ApoB mice were generated by crossing APP/PS1 transgenic mice with ApoB transgenic mice. Only male mice were used in our study, except during breeding, to reduce variability introduced by sex hormones and the estrous cycle in females. This approach enhances the consistency of behavioral and molecular outcomes, facilitating reproducible comparisons between experimental groups. All the mice were maintained in a specific pathogen-free (SPF) animal facility with a controlled environment (12-hour light/12-hour dark/day cycle, ~23°C, ~65% humidity). The mice were euthanized via CO₂ inhalation upon completion of the experiments. For the hypercholesterolemia mouse model, APP/PS1-ApoB mice were fed a high-fat high-cholesterol diet (HFHCD, Research Diets, NJ, USA, D12109C) at the age of 8 weeks. Following HFHCD feeding, the mice were subcutaneously administered a PCSK9 mAb (alirocumab, lot: EW2484, cat: S20190042, diluted to 7.5 mg/mL, administered 30 mg/kg mouse weight once every two weeks) or vehicle. Ethics approval All experiments on animal were conducted in accordance with the Declaration of Helsinki. Animal investigations were approved by the Institutional Animal Care and Use Committee of Fuwai Hospital, Chinese Academy of Medical Sciences (issue number: FW-2023-0033) and the Animal Ethics Committee of Cyagen Biosciences (approval numbers: AACU23-MS001-394 and AUC22-A463). Western blotting (WB) WB was performed by using previously described method[41,42]. proteins of mouse tissues were extracted by using 1% SDS cell lysis buffer (protease inhibitors included), quantified with a BCA kit (Pierce, Walham, MA, USA, 23225) and then electrophoresed and transferred. The anti-PCSK9 (1:1000, Abclonal, lot:4000000695, cat: A21909) and anti-GAPDH (1:1000, Proteintech, lot:01000415511mg, cat: 60004-1-Ig) antibodies were used. Morris Water Maze Visible Platform Test (Training Session - Day 1): The platform was raised above the water surface. Mice were sequentially placed into the water from four quadrants (northeast, northwest (NW), southeast (SE), and southwest) and guided to locate the visible platform. Spatial Navigation Test (Learning Assessment - Days 2-4): Over 3 consecutive days, the platform was submerged 1 cm below water surface. Each daily session consisted of four trials where mice were introduced into the maze facing the pool wall from one of four quadrants, with the platform consistently positioned in the SE quadrant. Each trial allowed 60 seconds of exploration for hidden platform localization. Animals that failed to locate the platform within 60 seconds were manually guided onto the platform and permitted equivalent 10-second residence times. Spatial probe test (Memory evaluation-Day 5): The platform was removed 24 hours after the final navigation test. Mice were introduced from the quadrant diametrically opposed to the original platform location (NW quadrant). Exploratory patterns were video recorded (Morris Water Maze Video Analysis System, Zhenghua Biological Instrument, Anhui, China, V2.0). Mice exhibiting minimal fluctuations in behavioral metrics throughout the training phase (i.e., no significant increase or decrease in target parameters) were excluded. Y-Maze The Y-maze apparatus (KEW BASIS, Nanjing, China, KW-Y) consisted of three symmetrical arms (labeled A, B, and C) angled at 120°. Mice were placed at the distal end of one randomly selected arm and allowed to freely explore the maze for 5 minutes. Total Arm Entries: The number of arm entries was quantified when all four paws crossed the arm threshold. Spontaneous Alternation: A complete alternation was defined as consecutive entries into all three distinct arms within a continuous sequence. Open Field Test Mice were individually placed in the central zone of an open-field arena (SHENRUIBIOTECH, Beijing, China, SR-OFBM02) under quiet conditions. Behavioral activity was recorded for 5 minutes using the VisuTrack video-tracking system to analyze locomotion and exploratory behavior. The following parameters were automatically quantified. Total Distance: Cumulative locomotor activity (m) across the entire arena. Center Entries: Number of transitions between predefined virtual grid sectors (25 × 25 cm subdivisions). Center Duration: Time (seconds) spent in the central zone (defined as the innermost 30% of the arena area), reflecting anxiety-related behavior. Rotarod Test Mice were arranged a 3-day training on a rotating rod (diameter: 3.5 cm). After that mice were acclimated for 30 seconds on the rotating rod before initiating the rotarod apparatus (Yiyan, Jinan, China, YLS-4D). The rod was accelerated to a constant speed of 30 revolutions per minute, and the trial duration was set to 300 seconds. Fall latency to was recorded as the time of the first fall from the rod. Nissl Staining Fixed brain tissues were sectioned (30 μm) and deparaffinized if embedded, then stained with 0.1% cresyl violet solution for 5–10 minutes. After brief rinsing in water, the sections were differentiated through graded ethanol (70% to 100%) and cleared in xylene before being covered with DPX mounting medium. This method selectively stains RNA-rich neuronal cell bodies purple, enabling visualization of cytoarchitecture via brightfield microscopy. Immunohistochemistry (IHC) For paraffin sections, deparaffinize and perform antigen retrieval. Block with 3% H₂O₂ and 5% BSA, then incubate with anti-β-amyloid (1-16) (Aβ1-16, 6E10, Biolegend, lot: B430350, cat: 803001) or anti-glial fibrillary acidic protein (GFAP, Abcam, lot: GR 3434957-1, cat: ab7260) primary antibody at 4°C overnight and HRP-conjugated secondary antibody. Mounting and imaging were performed via microscopy. Serum Biochemistry Test Serum total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C) were measured using an automatic biochemistry analyzer (Rayto, Shenzhen, China, Chemray 800). TC via the cholesterol oxidase/peroxidase assay (Rayto, S03042), HDL-C via selection suppression method (Rayto, S03025) and LDL-C via surfactant removal method (Rayto, S03029). non-HDL-C were calculated as TC minus HDL-C. Serum human ApoB contents were detected by an ELISA kit (FANKEW, Shanghai, China, F111497-A). Statistical Analyses of Animal Experiments Our pre-established exclusion criteria for behavioral data were as follows: data points exceed 3 standard deviations from the group mean were classified as statistical outliers; mice that fail to exhibit locomotor activity during cognitive tests; and subjects show no change in escape latency between the initial and final stages of training. The statistical analyses were performed using GraphPad Prism 10 software. The normally distributed data were tested by a two-sided unpaired t -test for two-group comparisons when equal variances were assumed, or Welch’s t test was performed. The non-normally distributed data were analyzed by Mann–Whitney U-tests for two-group comparisons. The significance level for all the tests was set at P < 0.05. Artificial intelligence (AI) and AI-assisted technologies in the writing process The authors state that AI and AI-assisted technologies were only used in the writing process to improve readability and language in accordance with the TlTAN Guidelines 2025[43]. The work has been reported in line with the ARRIVE criteria[44]. Declarations Author contributions: G.L., L.Z. and D.N. contributed to the conception of the study; D.N., D.L., and Z.Q. performed the experiment; D.N. and Z.F. contributed significantly to analysis and manuscript preparation; D.N. and Z.F. wrote the manuscript; G.L., L.Z. and Y.L. reviewed the manuscript. Funding: Dr. Guoping Li was awarded funding for this research from the Innovation Fund for Medical Sciences (CIFMS) of the Chinese Academy of Medical Sciences (CAMS) (2021-I2M-1-008) and the National High Level Hospital Clinical Research Funding (BJ-2024-219 and BJ-2023-237). Moreover, Dr. Guoping Li also received support from the National Natural Science Foundation of China (82370877 and 81970739). Competing interests: The authors declare that they have no conflict of interest. DATA AVAILABILITY Upon request, unique noncommercial reagents and source data can be obtained from the corresponding authors. References Sandesara PB, Virani SS, Fazio S, Shapiro MD. The forgotten lipids: Triglycerides, remnant cholesterol, and atherosclerotic cardiovascular disease risk. Endocr Rev 2019;40:537–57. https://doi.org/10.1210/er.2018-00184. Benziger CP, Stebbins A, Wruck LM, Effron MB, Marquis-Gravel G, Farrehi PM, et al. Aspirin dosing for secondary prevention of atherosclerotic cardiovascular disease in male and female patients: A secondary analysis of the ADAPTABLE randomized clinical trial. JAMA Cardiol 2024;9:808–16. https://doi.org/10.1001/jamacardio.2024.1712. Williams MS, Levine GN, Kalra D, Agarwala A, Baptiste D, Cigarroa JE, et al. 2025 AHA/ACC clinical performance and quality measures for patients with chronic coronary disease: A report of the American college of cardiology/american heart association joint committee on performance measures. J Am Coll Cardiol 2025:S0735-1097(25)282-7. https://doi.org/10.1016/j.jacc.2025.02.001. Vrints C, Andreotti F, Koskinas KC, Rossello X, Adamo M, Ainslie J, et al. 2024 ESC guidelines for the management of chronic coronary syndromes: Developed by the task force for the management of chronic coronary syndromes of the european society of cardiology (ESC) endorsed by the european association for cardio-thoracic surgery (EACTS). Eur Heart J 2024;45:3415–537. https://doi.org/10.1093/eurheartj/ehae177. Grundy SM, Stone NJ, Bailey AL, Beam C, Birtcher KK, Blumenthal RS, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of blood cholesterol: Executive summary: a report of the American college of cardiology/american heart association task force on clinical practice guidelines. Circulation 2019;139:e1046–81. https://doi.org/10.1161/CIR.0000000000000624. Zuliani G, Cavalieri M, Galvani M, Volpato S, Cherubini A, Bandinelli S, et al. Relationship between low levels of high-density lipoprotein cholesterol and dementia in the elderly. The InChianti study. J Gerontol A Biol Sci Med Sci 2010;65:559–64. https://doi.org/10.1093/gerona/glq026. Kim J-M, Stewart R, Shin I-S, Yoon J-S. Low cholesterol, cognitive function and alzheimer s disease in a community population with cognitive impairment. J Nutr, Health Aging 2002;6:320–3. Olmastroni E, Molari G, De Beni N, Colpani O, Galimberti F, Gazzotti M, et al. Statin use and risk of dementia or alzheimer’s disease: A systematic review and meta-analysis of observational studies. Eur J Prev Cardiol 2022;29:804–14. https://doi.org/10.1093/eurjpc/zwab208. Rajan KB, Mcaninch EA, Wilson RS, Dhana A, Evans-Lacko S, Evans DA. Statin initiation and risk of incident alzheimer disease and cognitive decline in genetically susceptible older adults. Neurology 2024;102:e209168. https://doi.org/10.1212/WNL.0000000000209168. Golomb BA, McGraw JJ, Evans MA, Dimsdale JE. Physician response to patient reports of adverse drug effects. Drug Saf 2007;30:669–75. https://doi.org/10.2165/00002018-200730080-00003. Sabatine MS, Giugliano RP, Wiviott SD, Raal FJ, Blom DJ, Robinson J, et al. Efficacy and safety of evolocumab in reducing lipids and cardiovascular events. N Engl J Med 2015;372:1500–9. https://doi.org/10.1056/NEJMoa1500858. Giugliano RP, Mach F, Zavitz K, Kurtz C, Im K, Kanevsky E, et al. Cognitive function in a randomized trial of evolocumab. N Engl J Med 2017;377:633–43. https://doi.org/10.1056/NEJMoa1701131. Sniderman AD, Dufresne L, Pencina KM, Bilgic S, Thanassoulis G, Pencina MJ. Discordance among apoB, non-high-density lipoprotein cholesterol, and triglycerides: Implications for cardiovascular prevention. Eur Heart J 2024;45:2410–8. https://doi.org/10.1093/eurheartj/ehae258. Johannesen CDL, Langsted A, Nordestgaard BG, Mortensen MB. Excess apolipoprotein B and cardiovascular risk in women and men. J Am Coll Cardiol 2024;83:2262–73. https://doi.org/10.1016/j.jacc.2024.03.423. Picard C, Nilsson N, Labonté A, Auld D, Rosa-Neto P, Alzheimer’s Disease Neuroimaging Initiative, et al. Apolipoprotein B is a novel marker for early tau pathology in alzheimer’s disease. Alzheimer’s Dement: J Alzheimer’s Assoc 2022;18:875–87. https://doi.org/10.1002/alz.12442. Löffler T, Flunkert S, Havas D, Sántha M, Hutter-Paier B, Steyrer E, et al. Impact of ApoB-100 expression on cognition and brain pathology in wild-type and hAPPsl mice. Neurobiol Aging 2013;34:2379–88. https://doi.org/10.1016/j.neurobiolaging.2013.04.008. De Oliveira J, Moreira ELG, De Bem AF. Beyond cardiovascular risk: Implications of familial hypercholesterolemia on cognition and brain function. Ageing Res Rev 2024;93:102149. https://doi.org/10.1016/j.arr.2023.102149. Soffer DE, Marston NA, Maki KC, Jacobson TA, Bittner VA, Peña JM, et al. Role of apolipoprotein B in the clinical management of cardiovascular risk in adults: An expert clinical consensus from the national lipid association. J Clin Lipidol 2024;18:e647–63. https://doi.org/10.1016/j.jacl.2024.08.013. Liu E, Zhang Y, Wang J-Z. Updates in alzheimer’s disease: From basic research to diagnosis and therapies. Transl Neurodegener 2024;13:45. https://doi.org/10.1186/s40035-024-00432-x. Parra Bravo C, Naguib SA, Gan L. Cellular and pathological functions of tau. Nat Rev Mol Cell Biol 2024;25:845–64. https://doi.org/10.1038/s41580-024-00753-9. Reiserer RS, Harrison FE, Syverud DC, McDonald MP. Impaired spatial learning in the APPSwe + PSEN1ΔE9 bigenic mouse model of alzheimer’s disease. Genes, Brain Behav 2007;6:54–65. https://doi.org/10.1111/j.1601-183X.2006.00221.x. Jonas MC, Costantini C, Puglielli L. PCSK9 is required for the disposal of non-acetylated intermediates of the nascent membrane protein BACE1. EMBO Rep 2008;9:916–22. https://doi.org/10.1038/embor.2008.132. Vilella A, Bodria M, Papotti B, Zanotti I, Zimetti F, Remaggi G, et al. PCSK9 ablation attenuates aβ pathology, neuroinflammation and cognitive dysfunctions in 5XFAD mice. Brain, Behavior, and Immunity 2024;115:517–34. https://doi.org/10.1016/j.bbi.2023.11.008. Purcell-Huynh DA, Farese RV, Johnson DF, Flynn LM, Pierotti V, Newland DL, et al. Transgenic mice expressing high levels of human apolipoprotein B develop severe atherosclerotic lesions in response to a high-fat diet. J Clin Invest 1995;95:2246–57. https://doi.org/10.1172/JCI117915. Tóth ME, Dukay B, Hoyk Z, Sántha M. Cerebrovascular changes and neurodegeneration related to hyperlipidemia: Characteristics of the human ApoB-100 transgenic mice. CPD 2020;26:1486–94. https://doi.org/10.2174/1381612826666200218101818. Cao Y-X, Sun D, Liu H-H, Jin J-L, Li S, Guo Y-L, et al. Improvement of definite diagnosis of familial hypercholesterolemia using an expanding genetic analysis. JACC: Asia 2021;1:82–9. https://doi.org/10.1016/j.jacasi.2021.04.001. Defesche JC, Gidding SS, Harada-Shiba M, Hegele RA, Santos RD, Wierzbicki AS. Familial hypercholesterolaemia. Nat Rev, Dis Primers 2017;3:17093. https://doi.org/10.1038/nrdp.2017.93. Evans SJ, Waller PC, Davis S. Use of proportional reporting ratios (PRRs) for signal generation from spontaneous adverse drug reaction reports. Pharmacoepidemiol Drug Saf 2001;10:483–6. https://doi.org/10.1002/pds.677. Eldjarn GH, Ferkingstad E, Lund SH, Helgason H, Magnusson OT, Gunnarsdottir K, et al. Large-scale plasma proteomics comparisons through genetics and disease associations. Nature 2023;622:348–58. https://doi.org/10.1038/s41586-023-06563-x. Kurki MI, Karjalainen J, Palta P, Sipilä TP, Kristiansson K, Donner KM, et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature 2023;613:508–18. https://doi.org/10.1038/s41586-022-05473-8. Boehm FJ, Zhou X. Statistical methods for mendelian randomization in genome-wide association studies: A review. Comput Struct Biotechnol J 2022;20:2338–51. https://doi.org/10.1016/j.csbj.2022.05.015. Verbanck M, Chen C-Y, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from mendelian randomization between complex traits and diseases. Nat Genet 2018;50:693–8. https://doi.org/10.1038/s41588-018-0099-7. Burgess S, Thompson SG. Erratum to: Interpreting findings from mendelian randomization using the MR-egger method. Eur J Epidemiol 2017;32:391–2. https://doi.org/10.1007/s10654-017-0276-5. Yavorska OO, Burgess S. MendelianRandomization: An R package for performing mendelian randomization analyses using summarized data. Int J Epidemiol 2017;46:1734–9. https://doi.org/10.1093/ije/dyx034. Assessment of causal effects of visceral adipose tissue on risk of cancers: a mendelian randomization study - PubMed n.d. https://pubmed.ncbi.nlm.nih.gov/35218345/ (accessed May 26, 2025). Jia Y, Guo D, Sun L, Shi M, Zhang K, Yang P, et al. Self-reported daytime napping, daytime sleepiness, and other sleep phenotypes in the development of cardiometabolic diseases: A mendelian randomization study. Eur J Prev Cardiol 2022;29:1982–91. https://doi.org/10.1093/eurjpc/zwac123. Greco M FD, Minelli C, Sheehan NA, Thompson JR. Detecting pleiotropy in mendelian randomisation studies with summary data and a continuous outcome. Stat Med 2015;34:2926–40. https://doi.org/10.1002/sim.6522. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: Effect estimation and bias detection through egger regression. Int J Epidemiol 2015;44:512–25. https://doi.org/10.1093/ije/dyv080. Rosoff DB, Bell AS, Jung J, Wagner J, Mavromatis LA, Lohoff FW. Mendelian randomization study of PCSK9 and HMG-CoA reductase inhibition and cognitive function. J Am Coll Cardiol 2022;80:653–62. https://doi.org/10.1016/j.jacc.2022.05.041. Li H, Zhang L, Yang F, Feng X, Fu R, Zhao R, et al. Lipid-lowering drugs affect lung cancer risk via sphingolipid metabolism: A drug-target mendelian randomization study. Front Genet 2023;14:1269291. https://doi.org/10.3389/fgene.2023.1269291. Ni D, Qi Z, Ma S, Wang Y, Liang D, Zhang X, et al. Membrane-associated ring–CH–type finger 2 protects against metabolic dysfunction-associated fatty liver disease by targeting fatty acid synthase. Molecular Metabolism 2025;96:102137. https://doi.org/10.1016/j.molmet.2025.102137. Liu C, Zhao H, Xiao S, Han T, Chen Y, Wang T, et al. Slx5p-Slx8p promotes accurate chromosome segregation by mediating the degradation of synaptonemal complex components during meiosis. Adv Sci (Weinh Baden-Wurtt Ger) 2020;7:1900739. https://doi.org/10.1002/advs.201900739. Agha R, Mathew G, Rashid R, Kerwan A, Al-Jabir A, Sohrabi C, et al. Transparency in the reporting of artificial INtelligence – the TITAN guideline. PJS 2025. https://doi.org/10.70389/PJS.100082. Kilkenny C, Browne WJ, Cuthill IC, Emerson M, Altman DG. Improving bioscience research reporting: The ARRIVE guidelines for reporting animal research. PLOS Biol 2010;8:e1000412. https://doi.org/10.1371/journal.pbio.1000412. Patel SB, Wyne KL, Afreen S, Belalcazar LM, Bird MD, Coles S, et al. American association of clinical endocrinology clinical practice guideline on pharmacologic management of adults with dyslipidemia. Endocr Pract 2025;31:236–62. https://doi.org/10.1016/j.eprac.2024.09.016. Guan Y, Liu X, Yang Z, Zhu X, Liu M, Du M, et al. PCSK9 promotes LDLR degradation by preventing SNX17-mediated LDLR recycling. Circulation 2025;151:1512–26. https://doi.org/10.1161/CIRCULATIONAHA.124.072336. Sabatine MS, Giugliano RP, Keech AC, Honarpour N, Wiviott SD, Murphy SA, et al. Evolocumab and clinical outcomes in patients with cardiovascular disease. N Engl J Med 2017;376:1713–22. https://doi.org/10.1056/NEJMoa1615664. Mannarino MR, Sahebkar A, Bianconi V, Serban M-C, Banach M, Pirro M. PCSK9 and neurocognitive function: Should it be still an issue after FOURIER and EBBINGHAUS results? J Clin Lipidol 2018;12:1123–32. https://doi.org/10.1016/j.jacl.2018.05.012. Feng Z, Li X, Tong WK, He Q, Zhu X, Xiang X, et al. Real-world safety of PCSK9 inhibitors: a pharmacovigilance study based on spontaneous reports in FAERS. Front Pharmacol 2022;13:894685. https://doi.org/10.3389/fphar.2022.894685. Koren MJ, Rodriguez F, East C, Toth PP, Watwe V, Abbas CA, et al. An “inclisiran first” strategy vs usual care in patients with atherosclerotic cardiovascular disease. J Am Coll Cardiol 2024;83:1939–52. https://doi.org/10.1016/j.jacc.2024.03.382. Wright RS, Koenig W, Landmesser U, Leiter LA, Raal FJ, Schwartz GG, et al. Safety and tolerability of inclisiran for treatment of hypercholesterolemia in 7 clinical trials. J Am Coll Cardiol 2023;82:2251–61. https://doi.org/10.1016/j.jacc.2023.10.007. Pearson GJ, Thanassoulis G, Anderson TJ, Barry AR, Couture P, Dayan N, et al. 2021 canadian cardiovascular society guidelines for the management of dyslipidemia for the prevention of cardiovascular disease in adults. Can J Cardiol 2021;37:1129–50. https://doi.org/10.1016/j.cjca.2021.03.016. Supplementary Files SupplementaryFigureS1S10.docx SupplementaryTableS1S5.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8520824","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":575357127,"identity":"ffe630fe-17b6-4a56-a1a5-116742fbfa92","order_by":0,"name":"Dongsheng Ni","email":"","orcid":"","institution":"Beijing Hospital","correspondingAuthor":false,"prefix":"","firstName":"Dongsheng","middleName":"","lastName":"Ni","suffix":""},{"id":575357128,"identity":"eec51133-020f-4c8d-85e6-177565e111ed","order_by":1,"name":"Zeying Feng","email":"","orcid":"","institution":"Central South University Third Xiangya Hospital","correspondingAuthor":false,"prefix":"","firstName":"Zeying","middleName":"","lastName":"Feng","suffix":""},{"id":575357129,"identity":"6b962245-2b58-4a91-8fce-920a05958d90","order_by":2,"name":"Dehuan Liang","email":"","orcid":"","institution":"Beijing Hospital","correspondingAuthor":false,"prefix":"","firstName":"Dehuan","middleName":"","lastName":"Liang","suffix":""},{"id":575357130,"identity":"17983ed0-e4ef-43ff-ac01-35a1dbed7573","order_by":3,"name":"Zhaolai Qi","email":"","orcid":"","institution":"Beijing Hospital","correspondingAuthor":false,"prefix":"","firstName":"Zhaolai","middleName":"","lastName":"Qi","suffix":""},{"id":575357131,"identity":"31f56231-cda9-4c50-8a29-f8e512afbdcb","order_by":4,"name":"Yao Lu","email":"","orcid":"","institution":"Central South University Third Xiangya Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yao","middleName":"","lastName":"Lu","suffix":""},{"id":575357132,"identity":"276f25e6-1a15-4895-8043-b83c1fd6b2f5","order_by":5,"name":"Liyong Zhu","email":"","orcid":"","institution":"Central South University Third Xiangya Hospital","correspondingAuthor":false,"prefix":"","firstName":"Liyong","middleName":"","lastName":"Zhu","suffix":""},{"id":575357133,"identity":"4d847b17-7f3b-4224-af96-ec5950691a16","order_by":6,"name":"Guoping Li","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIiWNgGAWjYBACPmYwJQHEBxsf/6iQkJMnpIUNroXxcLMxwxkLY8MGQlrgLObjbdKMbRWJDAcIaWFnvybxcYdFnrzbwTbpwnkSCYwNzA8f3cDrMJ4yyZlnJIoNzxxstp65TSKPnYHN2DgHv5Y0ad42icSNMw423uDdJlHM2MDDJk2clvkPGyR450gkNhwgqIX9GFjLfIaDTdK8DURp4WG2BPolcQPDwWbDGcckjA2bCfiFn//4wxsfd9Qlzm84/vDBh5o6OXn25oeP8WlhYOAxYGBsYGAwOAATYMarHATYH4C1yDcQVDkKRsEoGAUjFQAAgdJLRExaYTgAAAAASUVORK5CYII=","orcid":"","institution":"Beijing Hospital","correspondingAuthor":true,"prefix":"","firstName":"Guoping","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2026-01-05 11:34:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8520824/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8520824/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":100696808,"identity":"aee6ab81-c639-4888-8f48-299afeb73234","added_by":"auto","created_at":"2026-01-20 15:10:02","extension":"tif","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1098504,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/7bfa1a27c775364950b83577.tif"},{"id":100696381,"identity":"cd36cd3c-e695-421f-bd70-884865bce92c","added_by":"auto","created_at":"2026-01-20 15:04:17","extension":"tif","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1337596,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/992cd7b9c2528475366c2f9f.tif"},{"id":100696537,"identity":"91e945f1-3825-462d-9004-1dbb1eb64947","added_by":"auto","created_at":"2026-01-20 15:05:28","extension":"tif","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14586548,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/b8b453f475673df9b58bcdce.tif"},{"id":100696434,"identity":"4c45f86f-aa69-4114-ba5f-a299aef1def7","added_by":"auto","created_at":"2026-01-20 15:04:41","extension":"tif","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1698084,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/ab7431bfa085de8e297e4298.tif"},{"id":100696442,"identity":"8bf07d98-0521-4550-8ec1-49607b7de922","added_by":"auto","created_at":"2026-01-20 15:04:47","extension":"tif","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3583684,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/fd6dbea0daebe442a9b5fccd.tif"},{"id":100696615,"identity":"e057c95d-1f70-4c03-a87c-98352cecded3","added_by":"auto","created_at":"2026-01-20 15:07:03","extension":"tif","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10938588,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS10.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/9bac54a916d8a2cb7713325e.tif"},{"id":100696546,"identity":"530d12d2-6890-4a9d-8440-f5cb9f7fb783","added_by":"auto","created_at":"2026-01-20 15:05:39","extension":"tif","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1945328,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS9.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/604e12e329723adf3b1d2b7d.tif"},{"id":100696683,"identity":"c787b94d-8459-44a4-9950-7f185e9f8af6","added_by":"auto","created_at":"2026-01-20 15:08:06","extension":"xml","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12888,"visible":true,"origin":"","legend":"","description":"","filename":"jtrmJTRMD2600215.xml","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/1b55eb7f9e9a5808f9572898.xml"},{"id":100696428,"identity":"b8e2429d-9e93-4c4a-8afb-4cba7470449c","added_by":"auto","created_at":"2026-01-20 15:04:37","extension":"xml","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1318,"visible":true,"origin":"","legend":"","description":"","filename":"JTRMD2600215166684.go.xml","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/e84d67235fe193336627378b.xml"},{"id":100696633,"identity":"30833a0d-7813-480a-b7af-50b9d516f146","added_by":"auto","created_at":"2026-01-20 15:07:08","extension":"xml","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":845,"visible":true,"origin":"","legend":"","description":"","filename":"JTRMD2600215Import.xml","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/0c62f603268d1c7b61c96841.xml"},{"id":100696710,"identity":"c2786dd0-63e7-47c6-8b86-fbef4a458691","added_by":"auto","created_at":"2026-01-20 15:08:17","extension":"xml","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":102712,"visible":true,"origin":"","legend":"","description":"","filename":"JTRMD26002150enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/faea8f52786b7aa044f39971.xml"},{"id":100696705,"identity":"f4f976ae-03f6-4601-837f-c7ec2d5c956b","added_by":"auto","created_at":"2026-01-20 15:08:16","extension":"tif","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1098504,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/737d423aace5e4b42876ca33.tif"},{"id":100696631,"identity":"9d208f24-4114-480d-ad19-0f873046f64c","added_by":"auto","created_at":"2026-01-20 15:07:06","extension":"tif","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1337596,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/c9d2096f17557864bd14da4b.tif"},{"id":100696720,"identity":"24e3a9f6-e59c-4913-a43c-b4a4a447fd3e","added_by":"auto","created_at":"2026-01-20 15:08:28","extension":"tif","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14586548,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/acfc8e03a28c67c4009248a9.tif"},{"id":100696712,"identity":"c245425f-4f0e-460e-82b5-69aacc5a8df6","added_by":"auto","created_at":"2026-01-20 15:08:19","extension":"tif","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1698084,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/cb686293441e1de7c1144f53.tif"},{"id":100696736,"identity":"724f4ce8-28ce-4de6-be92-8e9c52e58efa","added_by":"auto","created_at":"2026-01-20 15:08:31","extension":"tif","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3583684,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/92b7c1cd176485bec3d0808f.tif"},{"id":100696563,"identity":"2b58e311-8db3-43ae-930a-19de73257370","added_by":"auto","created_at":"2026-01-20 15:06:24","extension":"tif","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10938588,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS10.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/7d76e6b728c59169507a4572.tif"},{"id":100696822,"identity":"135caa7e-5adb-4686-a9c6-8f9866c6b721","added_by":"auto","created_at":"2026-01-20 15:10:05","extension":"tif","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1945328,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS9.tif","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/7d12052c156afdbdaddc3383.tif"},{"id":100696657,"identity":"37377075-3626-4ca3-bb48-ab41fd21993f","added_by":"auto","created_at":"2026-01-20 15:07:31","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":52002,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/103fa84752f1fe2065a778de.png"},{"id":100696533,"identity":"e09b2440-4afb-4608-a64c-a104a5a0d04f","added_by":"auto","created_at":"2026-01-20 15:05:24","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":48527,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/c2379aadd8f4f507a642d898.png"},{"id":100696549,"identity":"906b1e69-85d9-4ff8-8390-1c8b70717e10","added_by":"auto","created_at":"2026-01-20 15:05:40","extension":"jpeg","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14586548,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/8f43b7b03970e8147fdbdc71.jpeg"},{"id":100696681,"identity":"fe7516ba-1415-427a-ab6b-8f642e900840","added_by":"auto","created_at":"2026-01-20 15:08:06","extension":"jpeg","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1698084,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/28be7d8128dc156dbe53a91f.jpeg"},{"id":100696540,"identity":"912e203e-5b18-485a-a2c4-228b88de6622","added_by":"auto","created_at":"2026-01-20 15:05:31","extension":"jpeg","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3583684,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/19a328cc48879f04b795a827.jpeg"},{"id":100696547,"identity":"99baf2ef-a1f6-4b05-835d-7c5b6311a25d","added_by":"auto","created_at":"2026-01-20 15:05:39","extension":"png","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":158555,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/6b5ae7cd1a43c075ef1b3a5c.png"},{"id":100696368,"identity":"68773b16-b310-41cd-8224-10418133088d","added_by":"auto","created_at":"2026-01-20 15:03:59","extension":"png","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":189833,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/fcade181d427560dec76eca7.png"},{"id":100696548,"identity":"60a278b3-0d2e-444d-9eac-398631b74201","added_by":"auto","created_at":"2026-01-20 15:05:39","extension":"png","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1926072,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/bb235e8f5829e992d1f7a3a4.png"},{"id":100696559,"identity":"81fbbff9-daaa-433f-9779-33003bee56b4","added_by":"auto","created_at":"2026-01-20 15:06:14","extension":"png","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":249948,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure4.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/3af84f98c9ef096c790ec7f7.png"},{"id":100696682,"identity":"652ecb6e-401e-42b3-9af2-bc87010d6423","added_by":"auto","created_at":"2026-01-20 15:08:06","extension":"png","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1512904,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure5.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/ebe08ed713d32cbbea1655a1.png"},{"id":100696439,"identity":"8b093264-d326-4f6d-9da5-b1785c51e2ac","added_by":"auto","created_at":"2026-01-20 15:04:44","extension":"png","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1425781,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigureS10.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/de69ee10ae333959623732f0.png"},{"id":100696545,"identity":"aa0837e5-5d82-44b0-8d10-1e4bc671decd","added_by":"auto","created_at":"2026-01-20 15:05:38","extension":"png","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":413657,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigureS9.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/01d7bb7631d16203d8e0c8ac.png"},{"id":100696531,"identity":"474c2d19-a6a7-4758-9f73-5cb8a0a4e7ca","added_by":"auto","created_at":"2026-01-20 15:05:23","extension":"png","order_by":34,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":32646,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/686052ef28e7d797144b9b6b.png"},{"id":100696717,"identity":"f15f5e99-2d90-44e6-b3cd-64b1b33556ac","added_by":"auto","created_at":"2026-01-20 15:08:26","extension":"png","order_by":35,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29975,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/d818cb819b544b47419a1600.png"},{"id":100696456,"identity":"bc733f22-d35f-4bf9-813c-0f70be988a7d","added_by":"auto","created_at":"2026-01-20 15:05:05","extension":"png","order_by":36,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1926072,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/9e1272d5cf132c8534b33b5a.png"},{"id":100696454,"identity":"5c784b0e-8fa2-47b0-9f92-34e329ad692e","added_by":"auto","created_at":"2026-01-20 15:05:04","extension":"png","order_by":37,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":249948,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/49d3cc5be109f921bd05c58e.png"},{"id":100696378,"identity":"8b934b7c-9819-4129-937c-943b94d3c011","added_by":"auto","created_at":"2026-01-20 15:04:15","extension":"png","order_by":38,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1512904,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/314c085041cb7f212ac2b7cc.png"},{"id":100696465,"identity":"9e99f26a-805d-4986-9692-b6187ffa9973","added_by":"auto","created_at":"2026-01-20 15:05:06","extension":"xml","order_by":39,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":97297,"visible":true,"origin":"","legend":"","description":"","filename":"JTRMD26002150structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/d7abeaad66d4eb5796adcec4.xml"},{"id":100696561,"identity":"78768c0c-64cf-449c-9906-3f642ed6e69d","added_by":"auto","created_at":"2026-01-20 15:06:17","extension":"html","order_by":40,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":115184,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/693f6d24f351288ad9c5ff4b.html"},{"id":100696530,"identity":"a8619515-b88d-4e9d-819a-ae2395d9018a","added_by":"auto","created_at":"2026-01-20 15:05:21","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1361072,"visible":true,"origin":"","legend":"\u003cp\u003eReporting Odds Ratio (ROR) and Information Component (IC) for cognitive adverse events associated with PCSK9 inhibitors and statins.\u003c/p\u003e","description":"","filename":"Figure1.tif.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/6ca5d675991d4e17e7c77c64.jpg"},{"id":100696376,"identity":"70597f00-9ae5-4a8b-9e24-e5376e16a174","added_by":"auto","created_at":"2026-01-20 15:04:09","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1449027,"visible":true,"origin":"","legend":"\u003cp\u003eForest plot of MR analysis of ApoB and PCSK9 on AD risk. Abbreviations: ApoB, apolipoprotein B; CI, confidence interval; MR, Mendelian randomization; OR, odds ratio; PCSK9, proprotein convertase subtilisin/kexin type 9.\u003c/p\u003e","description":"","filename":"Figure2.tif.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/1ee313aecea3a68dd4c5fdb3.jpg"},{"id":100696532,"identity":"4e6dca04-c734-41a2-afd8-083e3952c0cd","added_by":"auto","created_at":"2026-01-20 15:05:24","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":6402281,"visible":true,"origin":"","legend":"\u003cp\u003ePCSK9 does not impair learning or memory in mice. (A) A schematic diagram showing that 36-week-old APP/PS1-PCSK9HKO mice and their littermate APP/PS1-PCSK9F/F mice were sacrificed after cognitive performance tests. The diagram was created with FigDraw (authorization code: SUROTb6bb6). Heap maps of a mouse’s center point (B) and mean escape latency (C) for the entire duration of the spatial probe test in the water maze test. (B) The color scale represents the average search time. (C) n = 8‒9 mice/group; two-tailed unpaired Student’s \u003cem\u003et\u003c/em\u003e test. Heap maps of a mouse’s center point (D) and spontaneous alterations (E) during the Y maze test. (E) n = 12 mice/group, two-tailed unpaired Student’s \u003cem\u003et\u003c/em\u003e test. Nissl staining (F), Aβ\u003csup\u003e+\u003c/sup\u003e plaques (G) and GFAP\u003csup\u003e+\u003c/sup\u003e astrocytes (H) within the hippocampal regions of the mouse brain. Scale bar = 200 μm.\u003c/p\u003e","description":"","filename":"Figure3.tif.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/2d87de9fe4de0e3c6cc117cf.jpg"},{"id":100696560,"identity":"01f8f065-ca13-4015-b47b-6831d4c4387e","added_by":"auto","created_at":"2026-01-20 15:06:15","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":2149804,"visible":true,"origin":"","legend":"\u003cp\u003ePCSK9 does not alter nerve or protection behavior in mice. Movement trajectories of APP/PS1-PCSK9\u003csup\u003eHKO \u003c/sup\u003emice and their littermates in the open field test (A-D). (A) Plot showing a heat map of the animal's center point for the entire duration of the open field test, and the color scale represents the average search time. (B) Total distance traveled by the mice in the open field test. The center entry numbers (C) and time (seconds) spent (D) in the central zone within the open field test. The percentage of fallen mice (E) and number of falls (F) in the rotarod test. n = 12 mice/group for the open field test and rotarod test. Two-tailed unpaired Student’s \u003cem\u003et\u003c/em\u003e test for (B), the Mann‒Whitney U test for (C, F), Welch’s \u003cem\u003et\u003c/em\u003e test for (D) and the Gehan‒Breslow‒Wilcoxon test for (E) were performed.\u003c/p\u003e","description":"","filename":"Figure4.tif.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/341b7c7c556cbfb8d9b77e4a.jpg"},{"id":100696392,"identity":"3acc749c-9d86-4cad-aef1-d7f0a1e0f937","added_by":"auto","created_at":"2026-01-20 15:04:24","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1911195,"visible":true,"origin":"","legend":"\u003cp\u003ePCSK9 inhibitors do not increase cognitive risk. (A) Real-world pharmacovigilance study of ADRs associated with PCSK9 inhibitors. (B) Two-sample MR analysis assessing the causal effect of ApoB and/or PCSK9 on AD. (C) PCSK9 inhibition of gene-edited mice on cognitive function. The diagram was created via FigDraw (authorization code: RTPIW98fc4). ICSRs, individual case safety reports; FAERS, FDA Adverse Event Reporting System; SNP, single-nucleotide polymorphism; UKB, UK Biobank; AD, Alzheimer’s disease; HKO, hepatocyte-specific knockout; HFHCD, high-fat high-cholesterol diet.\u003c/p\u003e","description":"","filename":"Figure5.tif.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/1d0694b4f35bc11208baca9d.jpg"},{"id":102634889,"identity":"3706d7b2-937d-485e-a0e6-4f716f93650f","added_by":"auto","created_at":"2026-02-13 21:10:00","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":13952830,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/ce23c1c1-3f2a-4f6b-9ab1-576aac374c90.pdf"},{"id":100696539,"identity":"1e85b510-c41e-4e20-bc16-14071fd4451a","added_by":"auto","created_at":"2026-01-20 15:05:31","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":5344113,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigureS1S10.docx","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/f4a2cd208c5df896ab228eab.docx"},{"id":100696441,"identity":"c84cf403-3daf-4a6f-833d-59307839c664","added_by":"auto","created_at":"2026-01-20 15:04:45","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":48523,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTableS1S5.docx","url":"https://assets-eu.researchsquare.com/files/rs-8520824/v1/f9ec9ab7032e9c6a1fe9ebb7.docx"}],"financialInterests":"","formattedTitle":"PCSK9 Inhibitors Do Not Increase Cognitive Risk","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003e\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eAtherosclerotic cardiovascular disease (ASCVD) remains the leading cause of global morbidity and mortality[1,2], underscoring the imperative for aggressive dyslipidemia management[3\u0026ndash;5]. This has driven the development of progressively potent low-density lipoprotein (LDL) cholesterol-lowering therapies. However, given the indispensable role of cholesterol in myelin synthesis and neuronal membrane integrity, the pursuit of profound lipid reduction has simultaneously raised concerns regarding potential neurocognitive adverse events[6,7]. These concerns initially surfaced with statin therapies, often on the basis of observational data and spontaneous reports[8\u0026ndash;10]. Given the capacity of proprotein convertase subtilisin\u0026ndash;kexin type 9 (PCSK9) inhibitors to achieve even more profound reductions in atherogenic lipoproteins than statins do, the emergence of these agents has logically intensified such concerns. Indeed, while initial studies have begun to address this issue, the findings remain controversial. For example, the Open-Label Study of Long-Term Evaluation Against LDL-C (OSLER-1 and OSLER-2) trials noted a numerically higher incidence of adverse events with evolocumab than with standard therapy[11]. Conversely, dedicated neurocognitive trials such as EBBINGHAUS, which prospectively assessed cognitive changes as a primary endpoint, reported no adverse effects with evolocumab[12].\u003c/p\u003e\u003cp\u003eA critical limitation of existing neurocognitive safety studies is the general omission of baseline apolipoprotein B (ApoB) levels as a potential modulating factor. ApoB, the primary structural apolipoprotein of all atherogenic lipoproteins (LDL, very low-density lipoprotein, intermediate-density lipoprotein, and lipoprotein a), provides a direct measure of the total atherogenic particle burden[13,14]. This oversight is particularly salient given the emerging links between ApoB metabolism and cognitive function[15]. It is plausible that individuals with a high baseline ApoB burden\u0026mdash;representing a potentially \u0026lsquo;vulnerable brain\u0026rsquo; phenotype\u0026mdash;might be more susceptible to any subtle neurocognitive perturbations[16,17]. Furthermore, these patients often require more intensive or prolonged lipid lowering therapy (including PCSK9 inhibitors) to reach guideline-recommended targets, resulting in greater cumulative drug exposure[18]. Consequently, a compelling need exists to delineate the long-term cognitive impact of PCSK9 inhibitors, specifically in patients with high baseline ApoB concentrations.\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eHowever, prospectively investigating this topic in a dedicated large-scale, long-term randomized controlled trial (RCT) presents substantial logistical and ethical hurdles. Therefore, a strategy integrating evidence from diverse, complementary methodologies is warranted to assess potential risks in this vulnerable subgroup. In the present study, we aimed to synthesize and evaluate evidence from: 1) analysis of spontaneous adverse event reports from a global pharmacovigilance database; 2) Mendelian randomization (MR) studies examining genetically proxied PCSK9 inhibition stratified by genetic predisposition to elevated ApoB levels; and 3) findings from preclinical animal models mimicking high ApoB states with concomitant PCSK9 inhibition. By triangulating insights from these approaches, we sought to provide a more comprehensive assessment of the cognitive safety of PCSK9 inhibitors, particularly in the context of a preexisting high ApoB particle burden.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eDisproportionality analysis using the FAERS\u003c/h2\u003e \u003cp\u003eWe conducted a pharmacovigilance study of adverse drug reactions (ADRs) associated with PCSK9 inhibitors and statins using data from the FDA Adverse Event Reporting System (FAERS) database, a publicly accessible repository of safety reports submitted by patients, healthcare professionals, and pharmaceutical companies. A total of 55,357,463 individual case safety reports (ICSRs) were collected from the FAERS database between the first quarter of 2004 and the fourth quarter of 2024. The specific generic drug names for Statins are detailed in Supplementary Table\u0026nbsp;1. PCSK9 inhibitors include evolocumab and alirocumab. Adverse event list is provided in Supplementary Table\u0026nbsp;2. Among these, 525,251 reports listed statins as a suspected drug. Of these statin-associated reports, 229 were linked to the MedDRA High-Level Term (HLT) Alzheimer\u0026rsquo;s disease (AD) (incl subtypes)\u0026rsquo; and 383 to \u0026lsquo;Dementia (excl Alzheimer\u0026rsquo;s type)\u0026rsquo;. Similarly, 371,200 reports involved PCSK9 inhibitors as a suspected drug, with 36 reports corresponding to AD (incl subtypes)\u0026rsquo; and 167 to \u0026lsquo;Dementia (excl Alzheimer\u0026rsquo;s type)\u0026rsquo;.\u003c/p\u003e \u003cp\u003eIn our disproportionality analysis, no significant signal for an increased risk of cognitive adverse events was detected for PCSK9 inhibitors. For \u0026lsquo;Alzheimer\u0026rsquo;s disease (incl subtypes)\u0026rsquo;, the ROR was 0.62 (95% Credibility Interval (CI), 0.45\u0026ndash;0.87) and the corresponding Information Component (IC) was \u0026minus;\u0026thinsp;0.68 (95% Credibility Interval, -1.14 to -0.19). For \u0026lsquo;Dementia (excl Alzheimer\u0026rsquo;s type)\u0026rsquo;, the ROR was 0.92 (95% CI, 0.79\u0026ndash;1.08) with an IC of -0.11 (95% Credibility Interval, -0.33 to 0.11). Conversely, statin use was associated with a significant disproportionality signal for cognitive decline across both analytical methods. Specifically, the Reporting Odds Ratio (ROR) for \u0026lsquo;AD (incl subtypes)\u0026rsquo; with statin use was 2.86 (95% CI, 2.51\u0026ndash;3.26) with an IC of 1.49 (95% CI, 1.29 to 1.67), and for \u0026lsquo;Dementia (excl Alzheimer\u0026rsquo;s type)\u0026rsquo; the ROR was 1.51 (95% CI, 1.36\u0026ndash;1.67) with an IC of 0.58 (95% CI 0.43 to 0.73) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eMR analysis\u003c/h3\u003e\n\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eTwo-sample MR\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eSupplementary Table\u0026nbsp;3 lists the genetic markers employed for constructing ApoB and genetic PCSK9 scores. Supplementary table 4 presents comprehensive details concerning the final single nucleotide polymorphisms (SNPs) for ApoB and PCSK9, including effect and other alleles, beta coefficients, standard errors, and corresponding P-values. The F-statistics for the chosen SNPs varied between 21.52 and 28.67 for ApoB, and from 32.07 to 2288.71 for PCSK9, which suggests an absence of weak instrumental variable bias.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn the univariable two-sample MR analysis assessing the causal effect of ApoB on AD, the IVW method showed no statistically significant association [odds ratio (OR)\u0026thinsp;=\u0026thinsp;0.79, 95% CI\u0026thinsp;=\u0026thinsp;0.56 to 1.11, P\u0026thinsp;=\u0026thinsp;0.17] (Fig.\u0026nbsp;2). Consistent with this finding, supplementary MR approaches, including the weighted median, MR-Egger, and maximum likelihood methods, yielded similarly non-significant estimates (all P\u0026thinsp;\u0026gt;\u0026thinsp;0.05), suggesting a lack of robust evidence for a causal role of ApoB in AD risk. Meanwhile, the identical approaches utilized to explore the causal effect of PCSK9 on AD also did not yield statistically significant results as shown in Fig.\u0026nbsp;2 (all P\u0026thinsp;\u0026gt;\u0026thinsp;0.05).\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003e \u003cb\u003eFigure 2\u003c/b\u003e. Forest plot of MR analysis of ApoB and PCSK9 on AD risk. Abbreviations: ApoB, apolipoprotein B; CI, confidence interval; MR, Mendelian randomization; OR, odds ratio; PCSK9, proprotein convertase subtilisin/kexin type 9.\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eSensitivity analyses\u003c/h3\u003e\n\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe Cochran's Q test and MR-PRESSO analyses indicated no significant heterogeneity or outlying variants in ApoB. The Cochran's Q test indicated significant heterogeneity in PCSK9, yet MR-PRESSO failed to identify any abnormal SNPs. This observation suggests that the observed heterogeneity may be attributable to the weak pleiotropy of the wide distribution or minor bias in the multi-instrumental variables (Supplementary Table\u0026nbsp;5). The primary analyses were conducted employing a random-effects inverse variance-weighted (IVW) model, and the findings were consistent with robust methods (e.g., weighted median), thereby substantiating the reliability of the conclusions. The MR‒Egger regression intercept terms both exceeded the significance threshold (P\u0026thinsp;\u0026gt;\u0026thinsp;0.05), confirming minimal horizontal pleiotropic effects. Initial visual inspection of forest and funnel plots (Supplementary Figs.\u0026nbsp;1\u0026ndash;4) supported these findings. Additional leave-one-out sensitivity tests (Supplementary Figs.\u0026nbsp;5\u0026ndash;6) further confirmed the stability of our primary outcomes. For comprehensive evaluation, we implemented four distinct analytical approaches and generated scatter plots depicting AD associations (Supplementary Figs.\u0026nbsp;7\u0026ndash;8).\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003e2\u0026times;2 factorial MR analysis\u003c/em\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eOf the initial 501,938 participants from the UK Biobank, 2,653 were excluded due to a diagnosis of AD or stroke at or before baseline. An additional 141,23 individuals were excluded for missing genetic score data, and 97,782 were excluded due to incomplete covariate information. The final analytical sample comprised 387,380 participants, among whom 184,612 (47.66%) were male, with a median age of 57 years (Interquartile Range (IQR): 50\u0026ndash;63). Over a median follow-up of 13.64 years (IQR: 12.91\u0026ndash;14.35), a total of 2,914 incident AD cases were documented.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn the 2\u0026times;2 factorial MR analysis, using the \u0026ldquo;High genetic ApoB score \u0026amp; Low genetic PCSK9 score\u0026rdquo; group as the reference category to directly test our primary hypothesis, none of the comparison groups demonstrated statistically significant differences in AD risk (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). These findings align with our primary MR analyses, which similarly failed to support a causal role for either biomarker individually.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCombined Genetic Predisposition to ApoB and PCSK9 and Alzheimer\u0026rsquo;s Disease Risk\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHigh ApoB_PRS \u0026amp; Low PCSK9_PRS (HR, P)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLow ApoB_PRS \u0026amp; Low PCSK9_PRS (HR, P)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLow ApoB_PRS \u0026amp; High PCSK9_PRS (HR, P)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHigh ApoB_PRS \u0026amp; High PCSK9_PRS (HR, P)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCrude Model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eref, -\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.96 (0.87, 1.06), 0.45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.95 (0.86, 1.06), 0.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.97 (0.87, 1.07), 0.50\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel 1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eref, -\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.97 (0.88, 1.08), 0.60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.98 (0.88, 1.08), 0.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.98 (0.88, 1.08), 0.66\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eref, -\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.97 (0.88, 1.08), 0.60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.97 (0.88, 1.08), 0.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.98 (0.88, 1.08), 0.65\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel 3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eref, -\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.97 (0.88, 1.07), 0.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.97 (0.88, 1.08), 0.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.97 (0.88, 1.08), 0.61\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eModel 1: Adjusted for age, sex, ethnicity\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eModel 2: Adjusted for age, sex, ethnicity, Townsend deprivation index\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eModel 3: Adjusted for age, sex, ethnicity, Townsend deprivation index, smoking status, drinking status, exercise, BMI, hypertension, diabetes, dyslipidemia\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e\n\u003cp\u003eAbbreviations: AD = Alzheimer\u0026rsquo;s disease; ApoB = Apolipoprotein B; BMI = Body Mass Index; CI = Confidence Interval; HR = Hazard Ratio; PCSK9 = Proprotein Convertase Subtilisin/Kexin type 9; PRS = Polygenic Risk Score; Ref = Reference.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAnimal experiment 1: hepatocyte-specific knockout of PCSK9\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCognitive impairment mouse line\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo investigate whether the inhibition of mouse PCSK9 induces neurocognitive dysfunction, we designed two mouse models for validation.\u0026nbsp;APP/PS1, double transgenic mice expressing a chimeric mouse/human amyloid precursor protein (Mo/HuAPP695swe) and a mutant human presenilin 1 (PS1-dE9), is the most common neurodegenerative model in the early-onset AD [19,20]. Importantly, this mouse line begin to develop beta-Amyloid (A\u0026beta;) plaques and spatial memory deficit at 6-7 months of age[21]. Genetically, we bred hepatocyte-specific knockout (HKO) of PCSK9 based on the APP/PS1 mice. It\u0026rsquo;s suggested that PCSK9 in the brain may affect the cleavage of A\u0026beta; protein[22], and there are also reports that PCSK9 exacerbates the release of inflammatory factors from astrocytes[23]. Therefore, whole-body knockout of PCSK9 may introduce many interferences in determining whether peripheral PCSK9 intervention affects cognitive function, while PCSK9\u003csup\u003eHKO\u0026nbsp;\u003c/sup\u003ecan overcome this deficiency.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eLearning and memory\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe raised APP/PS1-PCSK9\u003csup\u003eHKO\u003c/sup\u003e mice under conventional breeding conditions until the age of 9 months to simulate long-term inhibition of PCSK9 (Figure 3A). Subsequently, we assessed whether long-term PCSK9 inhibition affected cognitive function using behavioral and neuropathological experiments (Figure 3A). The results of the\u0026nbsp;western blotting (WB) experiment showed that compared with the control mice (APP/PS1-PCSK9\u003csup\u003eF/F\u003c/sup\u003e), the expression of PCSK9 in the livers of APP/PS1-PCSK9\u003csup\u003eHKO\u003c/sup\u003e mice was significantly reduced, without affecting the expression in other tissues such as brain, kidney, heart, fat, and muscle (Supplementary Figure 9A). We also assessed the serum lipid profile alteration of hepatocyte-restricted PCSK9 deficiency in above mouse lines. Our results demonstrated that compared with control mice, APP/PS1-PCSK9\u003csup\u003eHKO\u003c/sup\u003e mice exhibited approximately 54% reductions in average serum LDL-C (Supplementary Figure 9B), 31% reductions in average total cholesterol (TC) (Supplementary Figure 9C) and 32% reductions in average non- high-density lipoprotein cholesterol (non-HDL-C) (Supplementary Figure 9D).\u003c/p\u003e\n\u003cp\u003eThe water maze results indicated that as the number of training days increased, escape latency taken by both APP/PS1-PCSK9\u003csup\u003eF/F\u003c/sup\u003e and APP/PS1-PCSK9\u003csup\u003eHKO\u003c/sup\u003e mice to locate the submerged platform gradually decreased, with no statistically significant differences between groups (Supplementary Figure 9E). During the probe trial of water maze (Figure 3B), the two groups of mice were no statistically significant differences in the escape latency (Figure 3C), the number of platform crossings (Supplementary Figure 9G), and the time spent in the target quadrant (Supplementary Figure 9H), with the similar average swimming speed (Supplementary Figure 9F).\u003c/p\u003e\n\u003cp\u003eThe Y-maze alternation test results showed, under the condition of the same total number of entries into the three arms of the Y-maze (Supplementary Figure 9I), there were no statistically significant differences in the proportion of consecutive entries into the three arms (Figure 3D, E) between the two groups of mice.\u003c/p\u003e\n\u003cp\u003eThe pathological staining results of the brain tissue of the mice indicated that hepatocyte-specific knockout of PCSK9, did not affect the size or number of Nissl bodies (Figure 3F), A\u0026beta;\u003csup\u003e+\u003c/sup\u003e plaques (Figure 3G), or GFAP\u003csup\u003e+\u003c/sup\u003e astrocytes (Figure 3H) in the hippocampal region.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNerve and protection behavior\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIn addition to direct assessment of learning and memory, we also examined cognition-related phenotypes, including locomotor activity, nerve-like behavior, and motor coordination. In the open field test (Figure 4A), APP/PS1-PCSK9\u003csup\u003eHKO\u003c/sup\u003e and APP/PS1-PCSK9\u003csup\u003eF/F\u003c/sup\u003e mice exhibited comparable numbers of entries (Figure 4C) into and times spent in the central zone (Figure 4D) while demonstrating similar total travel distances (Figure 4B). During the rotarod test, the proportion of mice falling from the rotating rod increased with escalating rotation speed in both APP/PS1-PCSK9\u003csup\u003eHKO\u003c/sup\u003e and APP/PS1-PCSK9\u003csup\u003eF/F\u003c/sup\u003e groups, with no significant intergroup differences observed in fall percentage (Figure 4E). Furthermore, the number of falls within the 300-second test period was also comparable \u0026nbsp;between groups (Figure 4F).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAnimal experiment 2:\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eneutralization of PCSK9 pharmacologically\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eHuman-\u003c/em\u003e\u003cem\u003elike lipid profile mouse line\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMice are naturally not susceptible to ASCVD, and one potential reason is that the serum lipid profile of mice differs significantly from that of humans, which is characterized mainly by a lower ratio of LDL to HDL fraction. This may lead to potential inconsistencies in the results of PCSK9 inhibition on cognitive function\u0026nbsp;between\u0026nbsp;mice and humans. However, it has been reported that ApoB transgenic mice can significantly increase the ratio of the LDL to HDL fraction, resulting in a human-like serum lipid composition[24,25]. Moreover, the ApoB gene is a common causative gene leading to hypercholesterolemia[26,27]. Third, increasing evidence has shown that ApoB is the most accurate marker of ASCVD risk [13]. Fourth, PCSK9 mAbs usually rapidly decrease elevated LDL-C, which is different from the findings in the PCSK9 knockout model. Considering these points, we designed a second mouse model to explore the impact of PCSK9 inhibition on cognitive function. We bred APP/PS1-ApoB triple transgenic mice, fed them a high-fat high-cholesterol diet (HFHCD) to induce hypercholesterolemia. And then administered a PCSK9 mAb to observe whether a rapid reduction in LDL-C would impair the cognitive function of APP/PS1-ApoB mice, which exhibited a human-like lipid profile (Supplementary Figure 10A).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCognitive\u0026nbsp;\u003c/em\u003e\u003cem\u003efunction\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eOur data revealed that the HFHCD increased mouse body weight (Supplementary Figure 10B). Then PCSK9 mAb sharply decreased the serum ApoB content (~66%; Supplementary Figure 10C), LDL-C content (~43%; Supplementary Figure 10D), TC content (~32%; Supplementary Figure 10E) and nonHDL-C content (~37%; Supplementary Figure 10F).\u0026nbsp;Further water maze (Supplementary Figure 10G-K), Y-maze (Supplementary Figure 10L-N), open field test (Supplementary Figure 10O-R), rotarod test (Supplementary Figure 10S-T) and brain tissue pathological staining (Supplementary Figure 10U-W) results revealed that pharmacological neutralization of PCSK9 did not alter locomotor activity, nerve-like behavior, motor coordination or\u0026nbsp;balance in APP/PS1-based mice.\u003c/p\u003e\n\u003cp\u003eTherefore, neither the hepatocyte-specific genetic knockout of PCSK9 nor its pharmacological neutralization adversely affected cognitive function in APP/PS1 mice.\u003c/p\u003e"},{"header":"DISCUSSION and CONCLUSION","content":"\u003cp\u003ePCSK9 is a powerful drug target for therapeutic strategies aimed at lowering LDL-C, which regulates LDLR turnover of hepatocytes[28,29]. However, the impact of PCSK9 on cognitive function remains controversial, as evidence from both clinical and preclinical studies is inconsistent [11,12,23,30]. PCSK9 is also known to influence central nervous system development, neuronal differentiation, and apoptosis[12,22,23]. Furthermore, the Open-Label Study of Long-Term Evaluation against LDL Cholesterol (OSLER) study revealed that neurocognitive events were reported more frequently in the evolocumab group an assessment of 11.1 months (median)[11]. These uncertainties highlight the need for rigorous investigation into the potential neurocognitive effects of prolonged PCSK9 inhibitor use.\u003c/p\u003e\n\u003cp\u003eTo address the issue of PCSK9 inhibitors on cognitive safety, we conducted the first study to systematically combine pharmacovigilance, genetic epidemiology, and preclinical evidence in a single investigation.\u003c/p\u003e\n\u003cp\u003eThe present study did not find evidence of adverse cognitive-related side effects related to the inhibition of PCSK9, in concurrence with earlier safety assessments based on short-term clinical studies and genetic studies[12,30\u0026ndash;32]. In our disproportionality analysis of FAERS data, no significant association was detected between the use of PCSK9 inhibitors and increased reporting of adverse events related to cognitive function. However, the use of statin has (Figure 1). To our knowledge, we provides the longest pharmacovigilance follow-up to date (~20 years)[33].\u003c/p\u003e\n\u003cp\u003eMR is a methodological approach in human genetics that facilitates causal inference. It effectively reduces the confounding bias frequently observed in conventional epidemiological\u0026nbsp;studies. When the assumptions underlying MR are rigorously met, the inherent randomization of genetic instruments offers a framework akin to a RCT, allowing for the assessment of long-term causal effects of exposures[30]. In the univariable two-sample MR analysis, we cannot conclude a causal relationship between PCSK9 and AD from\u0026nbsp;the\u0026nbsp;UKB\u0026nbsp;source by 4\u0026nbsp;MR analysis approaches\u0026nbsp;(Figure 2). Furthermore, our data also failed to support a causal role of combined genetic predisposition to ApoB and PCSK9 and AD risk by using 2\u0026times;2 factorial MR analysis (Table 1).\u003c/p\u003e\n\u003cp\u003eExcept for real world data and genetically proxied PCSK9 inhibition, we also developed two gene-edited animal models to verify whether the inhibition of PCSK9 impairs cognitive function. One is 9 mo APP/PS1-PCSK9\u003csup\u003eHKO\u003c/sup\u003e mice, which are used to simulate long-term inhibition of PCSK9 (Figure 3). The other group included PCSK9 mAb-administered APP/PS1-ApoB triple transgenic mice to simulate short-term inhibition of PCSK9 (Supplementary Figure 10). Our mouse behavior tests and brain pathological staining indicated that repression of PCSK9 did not exacerbate cognitive impairment, genetically or pharmacologically.\u003c/p\u003e\n\u003cp\u003eA recent study revealed that PCSK9 ablation attenuates A\u0026beta; pathology, neuroinflammation and cognitive dysfunctions in 5XFAD mice[23]. However, they used PCSK9 whole-body knockout mice, in which brain PCSK9 was also deleted. This differs from approved therapeutics that use PCSK9 mAbs (which were also employed in our animal experiments), as mAbs typically cannot cross the blood‒brain barrier to directly affect cerebral PCSK9 function. The currently approved PCSK9 siRNA drugs specifically target the liver and do not directly affect brain PCSK9 expression[34,35].\u003c/p\u003e\n\u003cp\u003eApoB combines LDL-C as a validated therapeutic target in dyslipidemia[36], although its neuroprotective threshold remains debated[16,25]. Our preclinical data confirmed that aggressive lowering of ApoB (mean reduction: 66%) in AD-relevant mouse models induced no detectable neurocognitive deficits (Supplementary Figure 10). For individuals with ApoB mutation-induced hypercholesterolemia, PCSK9 inhibitors do not pose additional cognitive risks when used to lower cholesterol (Figure 2 and Supplementary Figure 10). As far as we known, no previous study has examined this specific population using genetic and preclinical approaches.\u003c/p\u003e\n\u003cp\u003eOur study is subject to several limitations. First, the study populations were exclusively of European or American ancestry. Consequently, our findings may not be generalizable across all populations. Also, an APP/PS1 transgenic mouse model may not fully replicate human cognitive disease pathophysiology. Second, there were instances of heterogeneity within the MR analysis. The reliance on GWAS data precludes the exploration of potential nonlinear relationships or stratification effects that may vary according to age, health status, or sex. Finally, the animal experiments did not include female mice, which may limit the applicability of the results.\u003c/p\u003e\n\u003cp\u003eBy integrating real-world adverse drug reaction data, MR studies, and animal experimental results from this study, combined with existing related research, we conclude that, compared with statins, PCSK9 inhibitors do not increase the risk of neurocognitive impairment, especially given their cardiovascular benefits. This conclusion also applies to patients with hypercholesterolemia caused by ApoB mutations (Figure 5). Besides, these integrated approaches establish a new standard for drug safety evaluation that could be applied to other therapeutic areas.\u0026nbsp;\u003c/p\u003e"},{"header":"MATERIALS AND METHODS","content":"\u003cp\u003e\u003cstrong\u003eStudy Design\u003c/strong\u003e \u003c/p\u003e\n\u003cp\u003eTo address the issue of PCSK9 inhibitors on cognitive safety, pharmacovigilance analysis, MR analysis and preclinical gene-editing mice were conducted.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePharmacovigilance data analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFAERS data files spanning from the first quarter of 2004 to the fourth quarter of 2024 were downloaded for this analysis. Adverse event terms within the FAERS database were coded via the most current version of the Medical Dictionary for Regulatory Activities (MedDRA), version 27.1.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDrug \u003c/em\u003e\u003cem\u003eDefinitions\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe WHO Drug Dictionary was utilized to standardize all drug names within the database. Statins were defined via the Anatomical Therapeutic Chemical (ATC) classification code C10AA; the specific generic drug names included are detailed in Supplementary Table 1. PCSK9 inhibitors include evolocumab and alirocumab.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAdverse Event\u003c/em\u003e\u003cem\u003e Definition\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eCognitive-related adverse events were defined via standardized MedDRA terminology at the HLT level, encompassing the HLTs AD (incl subtypes)’ and ‘Dementia (excl Alzheimer’s type)’. A comprehensive list of the specific Preferred Terms (PTs) included under these HLTs is provided in Supplementary Table 2.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eStatistical analysis\u003c/em\u003e\u003cem\u003e in FAERS\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eDisproportionality analysis is a fundamental data mining and signal detection technique —a statistical method used to identify potential safety signals—in pharmacovigilance and pharmacoepidemiology aimed at identifying statistically significant safety signal associations between drugs and adverse events (AEs) within spontaneous reporting system databases. In this study, we employed the reporting odds ratio (ROR) and the Bayesian Confidence Propagation Neural Network (BCPNN) for disproportionality analysis to ensure methodological robustness. A signal was considered statistically significant if the lower limit of the 95% CI for the ROR was greater than 1. For BCPNN, a signal was identified if the lower bound of the 95% credibility interval for the IC was greater than 0. All analyses required that the number of ICSRs for the drug‒event combination was three or more[28].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMendelian randomization analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eGenome-wide association studies (GWAS) \u003c/em\u003e\u003cem\u003edata\u003c/em\u003e\u003cem\u003e sources\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe GWAS data of ApoB and PCSK9 were sourced from UKB (n = 46,218, aged 40 to 69 years old)[29]. The summary genetic data for AD were acquired from FinnGen (version R10)[30] including 10,520 cases and 401,661 controls of Finland descent (which are accessible at https://www.finngen.fi/en).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eConsent\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eEthical oversight for the UKB was provided by the Northwest Multi-Center Research Ethics Committee, with all study participants providing written consent. Our research team gained authorization to use these resources after receiving approval from the UKB Ethics and Governance Council (Application No. 75283).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSNP \u003c/em\u003e\u003cem\u003eselection\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo investigate the potential causal effects of ApoB and PCSK9 on AD susceptibility, rigorous instrumental variable selection criteria were implemented through multiple quality control steps. The initial stage involved identifying genetic variants exhibiting associations with the exposure variables, where single nucleotide polymorphisms (SNPs) demonstrating statistical significance with ApoB and PCSK9 were selected as candidate instruments. Due to the absence of genome-wide significant SNPs (P \u0026lt; 5 × 10\u003csup\u003e−8\u003c/sup\u003e) for ApoB, a more lenient threshold (P \u0026lt; 5 × 10\u003csup\u003e−6\u003c/sup\u003e) was adopted to ensure sufficient instrumental variables (IVs) for robust analysis. Subsequently, we clumped SNPs based on the basis of the European 1000 Genomes Project reference panel (r\u003csup\u003e2\u003c/sup\u003e \u0026lt; 0.01 and clump distance \u0026gt; 10000 kb) to guarantee genetic independence among selected variants. We applied a minor allele frequency threshold of \u0026gt; 0.01 to exclude rare variants and ensure genotype reliability. Data harmonization was then conducted to align effect alleles across exposure-outcome datasets. Finally, instrument strength was evaluated through F-statistics, with values less than 10 suggesting potential weak instrument bias and necessitating exclusion from further MR analysis.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTwo-\u003c/em\u003e\u003cem\u003esample\u003c/em\u003e\u003cem\u003e MR\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe employed multiple MR approaches, including inverse variance-weighted (IVW) regression, weighted median estimation, MR‒Egger regression, and maximum likelihood methods. The IVW method served as our primary analytical framework, operating under the assumption that all selected genetic variants meet instrumental variable assumptions[31]. The weighted median estimator provides robust causal inference when at least 50% of the weighting derives from valid instrumental variables[32]. MR‒Egger regression offers additional protection against directional pleiotropy by allowing all genetic variants to exhibit some pleiotropic effects while providing valid causal estimates[33]. For scenarios involving weak instrumental variables, the maximum likelihood method was implemented to ensure accurate confidence interval estimation[34]. Comprehensive methodological details for these analytical techniques are available in previously published literature [35,36].\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSensitivity analyses\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo ensure the robustness of our findings, we implemented multiple sensitivity analyses. First, we assessed heterogeneity across genetic variants via Cochran's Q test [37]. The MR-PRESSO method was subsequently applied to detect and correct for horizontal pleiotropy by identifying and removing outlier variants that might bias the causal estimates[32]. When the global test indicated significant pleiotropic effects (P \u0026lt; 0.05), we excluded the identified outliers and repeated the IVW analysis with the remaining SNPs. Second, we evaluated potential directional pleiotropy through MR‒Egger regression, interpreting an intercept term with P \u0026gt; 0.05 as evidence against substantial horizontal pleiotropy[38]. Finally, we conducted leave-one-out analyses to examine whether individual SNPs disproportionately influenced the overall results, thereby identifying potential pleiotropic effects attributable to specific genetic variants.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e2×2 factorial MR analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe present study utilized data obtained from the UKB, which requires the formal approval of research proposals prior to data access. Ethical oversight for the UKB was provided by the Northwest Multi-Center Research Ethics Committee, with all study participants providing written consent. Our research team gained authorization to use these resources after receiving approval from the UKB Ethics and Governance Council (Application No. 75283). The genetic markers employed for constructing ApoB and genetic PCSK9 scores are comprehensively listed in Supplementary Table 3, while the variant selection methodology has been previously detailed in published works[39,40]. Elevated scores correspond to increased genetic predisposition for elevated circulating protein concentrations.\u003c/p\u003e\n\u003cp\u003eWe performed 2×2 factorial MR analysis, excluding participants with an AD or stroke diagnosis at or before baseline and participants missing covariates. Then, we split them to 4 groups based on the median of genetic ApoB and PCSK9 score. Critically, to address our primary research hypothesis regarding the potential vulnerability of individuals with high baseline ApoB burden to neurocognitive effects from intensive PCSK9 inhibition, we specifically designated the “High genetic ApoB score \u0026amp; Low genetic PCSK9 score” group as our reference category. Our models were adjusted for age, sex, ethnicity, Townsend deprivation index, smoking status, drinking status, exercise, BMI, hypertension, diabetes, dyslipidemia.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAnimal experiments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMouse lines\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAPP/PS1 mouse (MMRRC Strain #034832-JAX), double transgenic mice expressing a chimeric mouse/human amyloid precursor protein (Mo/HuAPP695swe) and a mutant human presenilin 1 (PS1-dE9), were obtained from the Shanghai Model Organisms Center, Inc. The PCSK9\u003csup\u003eflox/flox\u003c/sup\u003e mouse strain (S-CKO-00068) and Alb-Cre (C001006) mouse strain, which expresses Cre recombinase under the albumin promoter, were purchased from Cyagen Biosciences (Suzhou, China). Tg(ApoB)1102Sgy N20+ mice (Model No. 1004, commonly referred to as ApoB mice), which exhibit high levels of human ApoB100, were purchased from Taconic Biosciences, Inc.\u003c/p\u003e\n\u003cp\u003eAPP/PS1-PCSK9\u003csup\u003eHKO\u003c/sup\u003e mice were generated by multiple crossings among APP/PS1 transgenic mice, PCSK9\u003csup\u003eFlox/Flox\u003c/sup\u003e mice and Alb-Cre mice. The resulting offspring with the genotype APP/PS1-PCSK9\u003csup\u003eF/F\u003c/sup\u003e-Alb-Cre\u003csup\u003e+ \u003c/sup\u003e(named APP/PS1-PCSK9\u003csup\u003eHKO\u003c/sup\u003e) exhibited hepatocyte-specific knockout (HKO) of PCSK9 and overexpression of human APP/PS1.\u003c/p\u003e\n\u003cp\u003eAPP/PS1-ApoB mice were generated by crossing APP/PS1 transgenic mice with ApoB transgenic mice.\u003c/p\u003e\n\u003cp\u003eOnly male mice were used in our study, except during breeding, to reduce variability introduced by sex hormones and the estrous cycle in females. This approach enhances the consistency of behavioral and molecular outcomes, facilitating reproducible comparisons between experimental groups. All the mice were maintained in a specific pathogen-free (SPF) animal facility with a controlled environment (12-hour light/12-hour dark/day cycle, ~23°C, ~65% humidity). The mice were euthanized via CO₂ inhalation upon completion of the experiments. For the hypercholesterolemia mouse model, APP/PS1-ApoB mice were fed a high-fat high-cholesterol diet (HFHCD, Research Diets, NJ, USA, D12109C) at the age of 8 weeks. Following HFHCD feeding, the mice were subcutaneously administered a PCSK9 mAb (alirocumab, lot: EW2484, cat: S20190042, diluted to 7.5 mg/mL, administered 30 mg/kg mouse weight once every two weeks) or vehicle.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eEthics approval\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAll experiments on animal were conducted in accordance with the Declaration of Helsinki. Animal investigations were approved by the Institutional Animal Care and Use Committee of Fuwai Hospital, Chinese Academy of Medical Sciences (issue number: FW-2023-0033) and the Animal Ethics Committee of Cyagen Biosciences (approval numbers: AACU23-MS001-394 and AUC22-A463).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eWestern blotting (WB)\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWB was performed by using previously described method[41,42]. proteins of mouse tissues were extracted by using 1% SDS cell lysis buffer (protease inhibitors included), quantified with a BCA kit (Pierce, Walham, MA, USA, 23225) and then electrophoresed and transferred. The anti-PCSK9 (1:1000, Abclonal, lot:4000000695, cat: A21909) and anti-GAPDH (1:1000, Proteintech, lot:01000415511mg, cat: 60004-1-Ig) antibodies were used.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMorris Water Maze\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eVisible Platform Test (Training Session - Day 1): The platform was raised above the water surface. Mice were sequentially placed into the water from four quadrants (northeast, northwest (NW), southeast (SE), and southwest) and guided to locate the visible platform. Spatial Navigation Test (Learning Assessment - Days 2-4): Over 3 consecutive days, the platform was submerged 1 cm below water surface. Each daily session consisted of four trials where mice were introduced into the maze facing the pool wall from one of four quadrants, with the platform consistently positioned in the SE quadrant. Each trial allowed 60 seconds of exploration for hidden platform localization. Animals that failed to locate the platform within 60 seconds were manually guided onto the platform and permitted equivalent 10-second residence times. Spatial probe test (Memory evaluation-Day 5): The platform was removed 24 hours after the final navigation test. Mice were introduced from the quadrant diametrically opposed to the original platform location (NW quadrant). Exploratory patterns were video recorded (Morris Water Maze Video Analysis System, Zhenghua Biological Instrument, Anhui, China, V2.0). Mice exhibiting minimal fluctuations in behavioral metrics throughout the training phase (i.e., no significant increase or decrease in target parameters) were excluded.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eY-Maze\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe Y-maze apparatus (KEW BASIS, Nanjing, China, KW-Y) consisted of three symmetrical arms (labeled A, B, and C) angled at 120°. Mice were placed at the distal end of one randomly selected arm and allowed to freely explore the maze for 5 minutes. Total Arm Entries: The number of arm entries was quantified when all four paws crossed the arm threshold. Spontaneous Alternation: A complete alternation was defined as consecutive entries into all three distinct arms within a continuous sequence.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eOpen Field Test\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMice were individually placed in the central zone of an open-field arena (SHENRUIBIOTECH, Beijing, China, SR-OFBM02) under quiet conditions. Behavioral activity was recorded for 5 minutes using the VisuTrack video-tracking system to analyze locomotion and exploratory behavior. The following parameters were automatically quantified. Total Distance: Cumulative locomotor activity (m) across the entire arena. Center Entries: Number of transitions between predefined virtual grid sectors (25 × 25 cm subdivisions). Center Duration: Time (seconds) spent in the central zone (defined as the innermost 30% of the arena area), reflecting anxiety-related behavior.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eRotarod Test\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMice were arranged a 3-day training on a rotating rod (diameter: 3.5 cm). After that mice were acclimated for 30 seconds on the rotating rod before initiating the rotarod apparatus (Yiyan, Jinan, China, YLS-4D). The rod was accelerated to a constant speed of 30 revolutions per minute, and the trial duration was set to 300 seconds. Fall latency to was recorded as the time of the first fall from the rod.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNissl Staining\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFixed brain tissues were sectioned (30 μm) and deparaffinized if embedded, then stained with 0.1% cresyl violet solution for 5–10 minutes. After brief rinsing in water, the sections were differentiated through graded ethanol (70% to 100%) and cleared in xylene before being covered with DPX mounting medium. This method selectively stains RNA-rich neuronal cell bodies purple, enabling visualization of cytoarchitecture via brightfield microscopy.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eImmunohistochemistry (IHC)\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFor paraffin sections, deparaffinize and perform antigen retrieval. Block with 3% H₂O₂ and 5% BSA, then incubate with anti-β-amyloid (1-16) (Aβ1-16, 6E10, Biolegend, lot: B430350, cat: 803001) or anti-glial fibrillary acidic protein (GFAP, Abcam, lot: GR 3434957-1, cat: ab7260) primary antibody at 4°C overnight and HRP-conjugated secondary antibody. Mounting and imaging were performed via microscopy.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSerum Biochemistry Test\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSerum total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C) were measured using an automatic biochemistry analyzer (Rayto, Shenzhen, China, Chemray 800). TC via the cholesterol oxidase/peroxidase assay (Rayto, S03042), HDL-C via selection suppression method (Rayto, S03025) and LDL-C via surfactant removal method (Rayto, S03029). non-HDL-C were calculated as TC minus HDL-C. Serum human ApoB contents were detected by an ELISA kit (FANKEW, Shanghai, China, F111497-A).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eStatistical Analyses of Animal Experiments\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eOur pre-established exclusion criteria for behavioral data were as follows: data points exceed 3 standard deviations from the group mean were classified as statistical outliers; mice that fail to exhibit locomotor activity during cognitive tests; and subjects show no change in escape latency between the initial and final stages of training. The statistical analyses were performed using GraphPad Prism 10 software. The normally distributed data were tested by a two-sided unpaired \u003cem\u003et\u003c/em\u003e-test for two-group comparisons when equal variances were assumed, or Welch’s \u003cem\u003et\u003c/em\u003e test was performed. The non-normally distributed data were analyzed by Mann–Whitney U-tests for two-group comparisons. The significance level for all the tests was set at \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eArtificial intelligence (AI) and AI-assisted technologies in the writing process\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors state that AI and AI-assisted technologies were only used in the writing process to improve readability and language in accordance with the TlTAN Guidelines 2025[43]. The work has been reported in line with the ARRIVE criteria[44].\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor contributions:\u003c/strong\u003e G.L., L.Z. and D.N. contributed to the conception of the study; D.N., D.L., and Z.Q. performed the experiment; D.N. and Z.F. contributed significantly to analysis and manuscript preparation; D.N. and Z.F. wrote the manuscript; G.L., L.Z. and Y.L. reviewed the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003e Dr. Guoping Li was awarded funding for this research from the Innovation Fund for Medical Sciences (CIFMS) of the Chinese Academy of Medical Sciences (CAMS) (2021-I2M-1-008) and the National High Level Hospital Clinical Research Funding (BJ-2024-219 and BJ-2023-237). Moreover, Dr. Guoping Li also received support from the National Natural Science Foundation of China (82370877 and 81970739).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests:\u0026nbsp;\u003c/strong\u003eThe authors declare that they have no conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDATA AVAILABILITY\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUpon request, unique noncommercial reagents and source data can be obtained from the corresponding authors.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSandesara PB, Virani SS, Fazio S, Shapiro MD. The forgotten lipids: Triglycerides, remnant cholesterol, and atherosclerotic cardiovascular disease risk. Endocr Rev 2019;40:537\u0026ndash;57. https://doi.org/10.1210/er.2018-00184.\u003c/li\u003e\n\u003cli\u003eBenziger CP, Stebbins A, Wruck LM, Effron MB, Marquis-Gravel G, Farrehi PM, et al. Aspirin dosing for secondary prevention of atherosclerotic cardiovascular disease in male and female patients: A secondary analysis of the ADAPTABLE randomized clinical trial. JAMA Cardiol 2024;9:808\u0026ndash;16. https://doi.org/10.1001/jamacardio.2024.1712.\u003c/li\u003e\n\u003cli\u003eWilliams MS, Levine GN, Kalra D, Agarwala A, Baptiste D, Cigarroa JE, et al. 2025 AHA/ACC clinical performance and quality measures for patients with chronic coronary disease: A report of the American college of cardiology/american heart association joint committee on performance measures. J Am Coll Cardiol 2025:S0735-1097(25)282-7. https://doi.org/10.1016/j.jacc.2025.02.001.\u003c/li\u003e\n\u003cli\u003eVrints C, Andreotti F, Koskinas KC, Rossello X, Adamo M, Ainslie J, et al. 2024 ESC guidelines for the management of chronic coronary syndromes: Developed by the task force for the management of chronic coronary syndromes of the european society of cardiology (ESC) endorsed by the european association for cardio-thoracic surgery (EACTS). Eur Heart J 2024;45:3415\u0026ndash;537. https://doi.org/10.1093/eurheartj/ehae177.\u003c/li\u003e\n\u003cli\u003eGrundy SM, Stone NJ, Bailey AL, Beam C, Birtcher KK, Blumenthal RS, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of blood cholesterol: Executive summary: a report of the American college of cardiology/american heart association task force on clinical practice guidelines. Circulation 2019;139:e1046\u0026ndash;81. https://doi.org/10.1161/CIR.0000000000000624.\u003c/li\u003e\n\u003cli\u003eZuliani G, Cavalieri M, Galvani M, Volpato S, Cherubini A, Bandinelli S, et al. Relationship between low levels of high-density lipoprotein cholesterol and dementia in the elderly. The InChianti study. J Gerontol A Biol Sci Med Sci 2010;65:559\u0026ndash;64. https://doi.org/10.1093/gerona/glq026.\u003c/li\u003e\n\u003cli\u003eKim J-M, Stewart R, Shin I-S, Yoon J-S. Low cholesterol, cognitive function and alzheimer s disease in a community population with cognitive impairment. J Nutr, Health Aging 2002;6:320\u0026ndash;3.\u003c/li\u003e\n\u003cli\u003eOlmastroni E, Molari G, De Beni N, Colpani O, Galimberti F, Gazzotti M, et al. Statin use and risk of dementia or alzheimer\u0026rsquo;s disease: A systematic review and meta-analysis of observational studies. Eur J Prev Cardiol 2022;29:804\u0026ndash;14. https://doi.org/10.1093/eurjpc/zwab208.\u003c/li\u003e\n\u003cli\u003eRajan KB, Mcaninch EA, Wilson RS, Dhana A, Evans-Lacko S, Evans DA. Statin initiation and risk of incident alzheimer disease and cognitive decline in genetically susceptible older adults. Neurology 2024;102:e209168. https://doi.org/10.1212/WNL.0000000000209168.\u003c/li\u003e\n\u003cli\u003eGolomb BA, McGraw JJ, Evans MA, Dimsdale JE. Physician response to patient reports of adverse drug effects. Drug Saf 2007;30:669\u0026ndash;75. https://doi.org/10.2165/00002018-200730080-00003.\u003c/li\u003e\n\u003cli\u003eSabatine MS, Giugliano RP, Wiviott SD, Raal FJ, Blom DJ, Robinson J, et al. Efficacy and safety of evolocumab in reducing lipids and cardiovascular events. N Engl J Med 2015;372:1500\u0026ndash;9. https://doi.org/10.1056/NEJMoa1500858.\u003c/li\u003e\n\u003cli\u003eGiugliano RP, Mach F, Zavitz K, Kurtz C, Im K, Kanevsky E, et al. Cognitive function in a randomized trial of evolocumab. N Engl J Med 2017;377:633\u0026ndash;43. https://doi.org/10.1056/NEJMoa1701131.\u003c/li\u003e\n\u003cli\u003eSniderman AD, Dufresne L, Pencina KM, Bilgic S, Thanassoulis G, Pencina MJ. Discordance among apoB, non-high-density lipoprotein cholesterol, and triglycerides: Implications for cardiovascular prevention. Eur Heart J 2024;45:2410\u0026ndash;8. https://doi.org/10.1093/eurheartj/ehae258.\u003c/li\u003e\n\u003cli\u003eJohannesen CDL, Langsted A, Nordestgaard BG, Mortensen MB. Excess apolipoprotein B and cardiovascular risk in women and men. J Am Coll Cardiol 2024;83:2262\u0026ndash;73. https://doi.org/10.1016/j.jacc.2024.03.423.\u003c/li\u003e\n\u003cli\u003ePicard C, Nilsson N, Labont\u0026eacute; A, Auld D, Rosa-Neto P, Alzheimer\u0026rsquo;s Disease Neuroimaging Initiative, et al. Apolipoprotein B is a novel marker for early tau pathology in alzheimer\u0026rsquo;s disease. Alzheimer\u0026rsquo;s Dement: J Alzheimer\u0026rsquo;s Assoc 2022;18:875\u0026ndash;87. https://doi.org/10.1002/alz.12442.\u003c/li\u003e\n\u003cli\u003eL\u0026ouml;ffler T, Flunkert S, Havas D, S\u0026aacute;ntha M, Hutter-Paier B, Steyrer E, et al. Impact of ApoB-100 expression on cognition and brain pathology in wild-type and hAPPsl mice. Neurobiol Aging 2013;34:2379\u0026ndash;88. https://doi.org/10.1016/j.neurobiolaging.2013.04.008.\u003c/li\u003e\n\u003cli\u003eDe Oliveira J, Moreira ELG, De Bem AF. Beyond cardiovascular risk: Implications of familial hypercholesterolemia on cognition and brain function. Ageing Res Rev 2024;93:102149. https://doi.org/10.1016/j.arr.2023.102149.\u003c/li\u003e\n\u003cli\u003eSoffer DE, Marston NA, Maki KC, Jacobson TA, Bittner VA, Pe\u0026ntilde;a JM, et al. Role of apolipoprotein B in the clinical management of cardiovascular risk in adults: An expert clinical consensus from the national lipid association. J Clin Lipidol 2024;18:e647\u0026ndash;63. https://doi.org/10.1016/j.jacl.2024.08.013.\u003c/li\u003e\n\u003cli\u003eLiu E, Zhang Y, Wang J-Z. Updates in alzheimer\u0026rsquo;s disease: From basic research to diagnosis and therapies. Transl Neurodegener 2024;13:45. https://doi.org/10.1186/s40035-024-00432-x.\u003c/li\u003e\n\u003cli\u003eParra Bravo C, Naguib SA, Gan L. Cellular and pathological functions of tau. Nat Rev Mol Cell Biol 2024;25:845\u0026ndash;64. https://doi.org/10.1038/s41580-024-00753-9.\u003c/li\u003e\n\u003cli\u003eReiserer RS, Harrison FE, Syverud DC, McDonald MP. Impaired spatial learning in the APPSwe + PSEN1\u0026Delta;E9 bigenic mouse model of alzheimer\u0026rsquo;s disease. Genes, Brain Behav 2007;6:54\u0026ndash;65. https://doi.org/10.1111/j.1601-183X.2006.00221.x.\u003c/li\u003e\n\u003cli\u003eJonas MC, Costantini C, Puglielli L. PCSK9 is required for the disposal of non-acetylated intermediates of the nascent membrane protein BACE1. EMBO Rep 2008;9:916\u0026ndash;22. https://doi.org/10.1038/embor.2008.132.\u003c/li\u003e\n\u003cli\u003eVilella A, Bodria M, Papotti B, Zanotti I, Zimetti F, Remaggi G, et al. PCSK9 ablation attenuates a\u0026beta; pathology, neuroinflammation and cognitive dysfunctions in 5XFAD mice. Brain, Behavior, and Immunity 2024;115:517\u0026ndash;34. https://doi.org/10.1016/j.bbi.2023.11.008.\u003c/li\u003e\n\u003cli\u003ePurcell-Huynh DA, Farese RV, Johnson DF, Flynn LM, Pierotti V, Newland DL, et al. Transgenic mice expressing high levels of human apolipoprotein B develop severe atherosclerotic lesions in response to a high-fat diet. J Clin Invest 1995;95:2246\u0026ndash;57. https://doi.org/10.1172/JCI117915.\u003c/li\u003e\n\u003cli\u003eT\u0026oacute;th ME, Dukay B, Hoyk Z, S\u0026aacute;ntha M. Cerebrovascular changes and neurodegeneration related to hyperlipidemia: Characteristics of the human ApoB-100 transgenic mice. CPD 2020;26:1486\u0026ndash;94. https://doi.org/10.2174/1381612826666200218101818.\u003c/li\u003e\n\u003cli\u003eCao Y-X, Sun D, Liu H-H, Jin J-L, Li S, Guo Y-L, et al. Improvement of definite diagnosis of familial hypercholesterolemia using an expanding genetic analysis. JACC: Asia 2021;1:82\u0026ndash;9. https://doi.org/10.1016/j.jacasi.2021.04.001.\u003c/li\u003e\n\u003cli\u003eDefesche JC, Gidding SS, Harada-Shiba M, Hegele RA, Santos RD, Wierzbicki AS. Familial hypercholesterolaemia. Nat Rev, Dis Primers 2017;3:17093. https://doi.org/10.1038/nrdp.2017.93.\u003c/li\u003e\n\u003cli\u003eEvans SJ, Waller PC, Davis S. Use of proportional reporting ratios (PRRs) for signal generation from spontaneous adverse drug reaction reports. Pharmacoepidemiol Drug Saf 2001;10:483\u0026ndash;6. https://doi.org/10.1002/pds.677.\u003c/li\u003e\n\u003cli\u003eEldjarn GH, Ferkingstad E, Lund SH, Helgason H, Magnusson OT, Gunnarsdottir K, et al. Large-scale plasma proteomics comparisons through genetics and disease associations. Nature 2023;622:348\u0026ndash;58. https://doi.org/10.1038/s41586-023-06563-x.\u003c/li\u003e\n\u003cli\u003eKurki MI, Karjalainen J, Palta P, Sipil\u0026auml; TP, Kristiansson K, Donner KM, et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature 2023;613:508\u0026ndash;18. https://doi.org/10.1038/s41586-022-05473-8.\u003c/li\u003e\n\u003cli\u003eBoehm FJ, Zhou X. Statistical methods for mendelian randomization in genome-wide association studies: A review. Comput Struct Biotechnol J 2022;20:2338\u0026ndash;51. https://doi.org/10.1016/j.csbj.2022.05.015.\u003c/li\u003e\n\u003cli\u003eVerbanck M, Chen C-Y, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from mendelian randomization between complex traits and diseases. Nat Genet 2018;50:693\u0026ndash;8. https://doi.org/10.1038/s41588-018-0099-7.\u003c/li\u003e\n\u003cli\u003eBurgess S, Thompson SG. Erratum to: Interpreting findings from mendelian randomization using the MR-egger method. Eur J Epidemiol 2017;32:391\u0026ndash;2. https://doi.org/10.1007/s10654-017-0276-5.\u003c/li\u003e\n\u003cli\u003eYavorska OO, Burgess S. MendelianRandomization: An R package for performing mendelian randomization analyses using summarized data. Int J Epidemiol 2017;46:1734\u0026ndash;9. https://doi.org/10.1093/ije/dyx034.\u003c/li\u003e\n\u003cli\u003eAssessment of causal effects of visceral adipose tissue on risk of cancers: a mendelian randomization study - PubMed n.d. https://pubmed.ncbi.nlm.nih.gov/35218345/ (accessed May 26, 2025).\u003c/li\u003e\n\u003cli\u003eJia Y, Guo D, Sun L, Shi M, Zhang K, Yang P, et al. Self-reported daytime napping, daytime sleepiness, and other sleep phenotypes in the development of cardiometabolic diseases: A mendelian randomization study. Eur J Prev Cardiol 2022;29:1982\u0026ndash;91. https://doi.org/10.1093/eurjpc/zwac123.\u003c/li\u003e\n\u003cli\u003eGreco M FD, Minelli C, Sheehan NA, Thompson JR. Detecting pleiotropy in mendelian randomisation studies with summary data and a continuous outcome. Stat Med 2015;34:2926\u0026ndash;40. https://doi.org/10.1002/sim.6522.\u003c/li\u003e\n\u003cli\u003eBowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: Effect estimation and bias detection through egger regression. Int J Epidemiol 2015;44:512\u0026ndash;25. https://doi.org/10.1093/ije/dyv080.\u003c/li\u003e\n\u003cli\u003eRosoff DB, Bell AS, Jung J, Wagner J, Mavromatis LA, Lohoff FW. Mendelian randomization study of PCSK9 and HMG-CoA reductase inhibition and cognitive function. J Am Coll Cardiol 2022;80:653\u0026ndash;62. https://doi.org/10.1016/j.jacc.2022.05.041.\u003c/li\u003e\n\u003cli\u003eLi H, Zhang L, Yang F, Feng X, Fu R, Zhao R, et al. Lipid-lowering drugs affect lung cancer risk via sphingolipid metabolism: A drug-target mendelian randomization study. Front Genet 2023;14:1269291. https://doi.org/10.3389/fgene.2023.1269291.\u003c/li\u003e\n\u003cli\u003eNi D, Qi Z, Ma S, Wang Y, Liang D, Zhang X, et al. Membrane-associated ring\u0026ndash;CH\u0026ndash;type finger 2 protects against metabolic dysfunction-associated fatty liver disease by targeting fatty acid synthase. Molecular Metabolism 2025;96:102137. https://doi.org/10.1016/j.molmet.2025.102137.\u003c/li\u003e\n\u003cli\u003eLiu C, Zhao H, Xiao S, Han T, Chen Y, Wang T, et al. Slx5p-Slx8p promotes accurate chromosome segregation by mediating the degradation of synaptonemal complex components during meiosis. Adv Sci (Weinh Baden-Wurtt Ger) 2020;7:1900739. https://doi.org/10.1002/advs.201900739.\u003c/li\u003e\n\u003cli\u003eAgha R, Mathew G, Rashid R, Kerwan A, Al-Jabir A, Sohrabi C, et al. Transparency in the reporting of artificial INtelligence \u0026ndash; the TITAN guideline. PJS 2025. https://doi.org/10.70389/PJS.100082.\u003c/li\u003e\n\u003cli\u003eKilkenny C, Browne WJ, Cuthill IC, Emerson M, Altman DG. Improving bioscience research reporting: The ARRIVE guidelines for reporting animal research. PLOS Biol 2010;8:e1000412. https://doi.org/10.1371/journal.pbio.1000412.\u003c/li\u003e\n\u003cli\u003ePatel SB, Wyne KL, Afreen S, Belalcazar LM, Bird MD, Coles S, et al. American association of clinical endocrinology clinical practice guideline on pharmacologic management of adults with dyslipidemia. Endocr Pract 2025;31:236\u0026ndash;62. https://doi.org/10.1016/j.eprac.2024.09.016.\u003c/li\u003e\n\u003cli\u003eGuan Y, Liu X, Yang Z, Zhu X, Liu M, Du M, et al. PCSK9 promotes LDLR degradation by preventing SNX17-mediated LDLR recycling. Circulation 2025;151:1512\u0026ndash;26. https://doi.org/10.1161/CIRCULATIONAHA.124.072336.\u003c/li\u003e\n\u003cli\u003eSabatine MS, Giugliano RP, Keech AC, Honarpour N, Wiviott SD, Murphy SA, et al. Evolocumab and clinical outcomes in patients with cardiovascular disease. N Engl J Med 2017;376:1713\u0026ndash;22. https://doi.org/10.1056/NEJMoa1615664.\u003c/li\u003e\n\u003cli\u003eMannarino MR, Sahebkar A, Bianconi V, Serban M-C, Banach M, Pirro M. PCSK9 and neurocognitive function: Should it be still an issue after FOURIER and EBBINGHAUS results? J Clin Lipidol 2018;12:1123\u0026ndash;32. https://doi.org/10.1016/j.jacl.2018.05.012.\u003c/li\u003e\n\u003cli\u003eFeng Z, Li X, Tong WK, He Q, Zhu X, Xiang X, et al. Real-world safety of PCSK9 inhibitors: a pharmacovigilance study based on spontaneous reports in FAERS. Front Pharmacol 2022;13:894685. https://doi.org/10.3389/fphar.2022.894685.\u003c/li\u003e\n\u003cli\u003eKoren MJ, Rodriguez F, East C, Toth PP, Watwe V, Abbas CA, et al. An \u0026ldquo;inclisiran first\u0026rdquo; strategy vs usual care in patients with atherosclerotic cardiovascular disease. J Am Coll Cardiol 2024;83:1939\u0026ndash;52. https://doi.org/10.1016/j.jacc.2024.03.382.\u003c/li\u003e\n\u003cli\u003eWright RS, Koenig W, Landmesser U, Leiter LA, Raal FJ, Schwartz GG, et al. Safety and tolerability of inclisiran for treatment of hypercholesterolemia in 7 clinical trials. J Am Coll Cardiol 2023;82:2251\u0026ndash;61. https://doi.org/10.1016/j.jacc.2023.10.007.\u003c/li\u003e\n\u003cli\u003ePearson GJ, Thanassoulis G, Anderson TJ, Barry AR, Couture P, Dayan N, et al. 2021 canadian cardiovascular society guidelines for the management of dyslipidemia for the prevention of cardiovascular disease in adults. Can J Cardiol 2021;37:1129\u0026ndash;50. https://doi.org/10.1016/j.cjca.2021.03.016.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8520824/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8520824/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eProprotein convertase subtilisin\u0026ndash;kexin type 9 (PCSK9) inhibitors are potent lipid-lowering therapies, yet concerns regarding their neurocognitive safety persist amidst conflicting evidence.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe applied an integrated approach combining real-world pharmacovigilance, genetic epidemiology, and preclinical models. Disproportionality analysis was performed using FDA Adverse Event Reporting System (FAERS) data to assess cognitive adverse events. Two-sample Mendelian randomization (MR) was used to analyze the causal effects of apolipoprotein B (ApoB)-lowering and PCSK9 inhibition on Alzheimer\u0026rsquo;s disease (AD) via UK Biobank/FinnGen data, with a factorial MR design to evaluate interactions stratified by ApoB/PCSK9 polygenic risk. Preclinically, hepatocyte-specific knockout and pharmacological inhibition of PCSK9 in AD-transgenic mice were assessed through cognitive testing and neuropathological evaluations.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eNo increased risk signal for cognitive events was identified with PCSK9 inhibitors in FAERS analysis (AD Reporting Odds Ratio [ROR]\u0026thinsp;=\u0026thinsp;0.62, 95% CI 0.45\u0026ndash;0.87; dementia ROR\u0026thinsp;=\u0026thinsp;0.92, 0.79\u0026ndash;1.08), unlike statins which showed a significant signal (AD ROR\u0026thinsp;=\u0026thinsp;2.86). MR analysis revealed no causal association between PCSK9 inhibition (OR\u0026thinsp;=\u0026thinsp;1.00, 0.93\u0026ndash;1.07) or its ApoB-lowering effects (OR\u0026thinsp;=\u0026thinsp;0.79, 0.56\u0026ndash;1.11) and AD. In AD-transgenic mice, PCSK9 knockout reduced low-density lipoprotein-cholesterol by 54% without impairing cognition or neuron integrity, or increasing amyloid-β plaques and astrocytes. Pharmacological inhibition of PCSK9 similarly showed no cognitive deficits.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eIntegrated evidence from pharmacovigilance, genetic epidemiology, and preclinical models robustly supports the neurocognitive safety of PCSK9 inhibitors, including in individuals with high ApoB levels. These findings affirm their long-term neurocognitive safety profile in the management of atherosclerotic cardiovascular disease.\u003c/p\u003e","manuscriptTitle":"PCSK9 Inhibitors Do Not Increase Cognitive Risk","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-20 12:30:00","doi":"10.21203/rs.3.rs-8520824/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"72fb701b-a3e4-43a0-8008-f6c298a14a6a","owner":[],"postedDate":"January 20th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-05T17:53:35+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-20 12:30:00","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8520824","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8520824","identity":"rs-8520824","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}